Physical AI Brief

AI agents that carry a multi-step computer task through on their own became ordinary tools in the past year, and the same autonomy is available to anyone whose task is harmful. We ask what that means for a relying party -- an insurer, a lender, an auditor -- whose evidence is a filed PDF. AgentForge-Bench measures how reliably an off-the-shelf coding agent, driving one of seven open-weight models with a shell and the stock Python PDF stack, alters one dollar amount, date or address in a real filed financial document from a single sentence of intent, graded by rules rather than by a model. Across 1,750 cells, 1,419 (81.1%) satisfy the verifier, and 808 (46.2%) also survive every stricter filter: visible, localized, typeface-matched, original value gone document-wide. A deterministic script with no model in it solves 98 of the 125 documents; the agents solve 124, and none the script solves alone. Agents misreport 41% of their wrong edits as done, no model refused, and the cheapest verified forgery costs 2.4 cents. The raw rate overstates the threat by about a factor of two; the strict rate is still large.

Cardiovascular magnetic resonance (CMR) provides comprehensive cardiac assessment but remains underutilized because of the complexity of acquisition, post-processing, and interpretation. Existing artificial intelligence (AI) methods address isolated tasks, limiting clinical integration. We present ORION-CMR (On-scanner Reporting with Integrated fOunda-tioN Model), the first clinically evaluated scanner-native end-to-end CMR foundation model. Pretrained on 12,896,733 CMR images from 9,258 studies, ORION-CMR performs sequence classification, ventricular function assessment, late gadolinium enhancement (LGE) detection, binary and multiclass disease classification, and local large language model-based report generation in approximately 90 seconds. The framework. was evaluated on public benchmarks and clinically validated in a multi-vendor cohort of 68 subjects with normal examinations, congenital heart disease, dilated cardiomyopathy, and myocardial infarction. ORION-CMR outperformed supervised baselines and the previously published CMR foundation model (CMR-FM), achieving state-of-the-art performance for LGE classification and scar segmentation. Clinical evaluation achieved an AUC of 0.96 for normal-versus abnormal classification and 0.88 for multiclass disease classification, while generated reports demonstrated 81.4% agreement with expert interpretation. These results demonstrate the feasibility of real-time scanner-native AI-assisted CMR analysis and automated report generation.

In-band full-duplex (IBFD) multiple-input multiple-output (MIMO) systems enable simultaneous transmission and reception on the same frequency band, improving spectral efficiency for next-generation wireless networks. However, IBFD-MIMO systems are susceptible to self-interference (SI), which may overpower signals of interest (SOI). In this scenario, blind source separation (BSS) algorithms can be adopted to remove SI and perform joint sensing and communication (JSAC), but BSS algorithms mostly assume an idealized linear and quasi-stationary signal model, which does not hold under realistic radio frequency (RF) impairments, such as I/Q imbalance, carrier frequency offset (CFO), phase noise, and power amplifier nonlinearity. This paper proposes a two-stage echo state network (ESN)-based scheme that is superior to BSS under these realistic conditions. A frozen ESN is trained offline to characterize the static SI path, while an adaptive ESN, updated online via recursive least squares, tracks the time-varying SOI path using sparse pilot symbols. We evaluate the proposed scheme's SOI recovery performance and acquisition speed with different block sizes, comparing it against other recurrent neural networks (RNN), such as long short-term memory (LSTM) and gated recurrent unit (GRU). Simulation results show that the proposed approach outperforms BSS, LSTM, and GRU in both efficiency and SOI recovery, demonstrating the viability of ESNs for real-time, nonlinear self-interference cancellation in realistic IBFD MIMO systems.

Vision-language-action (VLA) policies can struggle with manipulation tasks with complex obstacle geometries due to partial observability. These complex geometries can lead to similar visual observations or robot configurations requiring qualitatively different actions, a distinction that can be quantified using topological signatures. While motion planners with full knowledge of environment geometries and object states can reason about these signatures in planning, this information is often not known at deployment. To address this issue, we present a topology-guided visual-prompting framework that uses simulation-based planning to augment a nominal demonstration dataset and provides vision-based guidance at deployment. Our method uses a Gauss-Linking-Integral topological signature representation to capture important topological properties of the environment. Using privileged geometry information from a simulation approximation of our environment, we augment a VLA fine-tuning dataset with trajectories that move the system to a demonstrated signature and, from the new configuration, resume task execution. A vision-language model (VLM) is fine-tuned on the same dataset to both predict signatures from live camera observations and predict end-effector waypoints, which are rendered as visual prompts on the observations to guide the VLA. Across three simulated bimanual tasks and a real-world box pickup task, our method outperforms a VLA fine-tuned only on nominal demonstrations and a VLM-prompting baseline that can remove topology-relevant information from observations. On hardware, it exceeds the strongest baseline by 40% in task success. Project website: https://topology-vla.github.io.

Underwater robot learning relies on simulators that integrate high-fidelity hydrodynamics, convenient learning interfaces, and a credible transition to real scenarios. In this work, we present FinsSim, a reality-aligned integrated simulation platform for Sim-to-Real underwater robot learning. FinsSim first constructs high-fidelity simulation with selectable backends to adapt to diverse requirements. To facilitate underwater robot research, it further offers standard control baselines, alongside with unified robot learning workflows. For reliable Sim-to-Real transfer, FinsSim adopts a multi-sensor fusion scheme to provide low-cost yet precise localization. Moreover, it implements calibrated thruster-hydrodynamics models and a constrained wrench allocation algorithm. Bridging these modules by ROS~2, FinsSim establishes a complete Sim-to-Real transfer pipeline. Through matched simulations and experiments, it is demonstrated that reliable Sim-to-Real transfer of underwater robot control policies can be achieved with the FinsSim framework. Separate ablation studies also validate that the modules of FinsSim can address the pivotal issues of underwater Sim-to-Real from different aspects. Overall, this work aims to bridge the gap between theoretical research and practical applications, ultimately driving advancements in the field of underwater robotics.

Effective communication between users and AI agents is essential for human-AI collaboration. The XY problem is a well-known communication pitfall where a person asks about their attempted solution rather than their actual problem. We extend prior sycophancy evaluation to the XY problem in agentic settings, evaluating whether agents can resist plausible but misleading suggestions from users and communicate their reasoning. We introduce XYEval, a meta-evaluation framework that can transform an existing benchmark into an XY problem evaluation. We evaluate five models across six diverse benchmark suites. Agents suffer large XY drops under XY mutation across benchmarks, with relative drops reaching up to 46.7%. With $τ^2$-bench, we further show that agent performance drops more when encountering a pedantic user who requires detailed explanations before approving a better solution. Our findings suggest that current agents lack the ability to effectively reason and communicate when facing misleading suggestions. A simple system instruction baseline that encourages awareness of XY problems only offers partial mitigation. Extensive trace analyses provide behavioral insights into how and why these XY drops occur across execution trajectories. Our results show that mitigating the XY problem remains challenging, requiring agents to both recognize user misdirection and clearly communicate the underlying problem.

Post-training turns a general next-token predictor into a chat model with a persistent assistant persona. If that persona is a character the model plays only on its own turns, its preferences should govern what the assistant says, not what the model predicts other speakers will say. We test this boundary and find that it does not hold: a safety-relevant preference of the assistant---for harmless over harmful tasks---shapes the model's predictions even on the user's turn, where the assistant is not the one speaking. We find that this preference is small or near-zero in pretrained base models, that it emerges through post-training, replicated across open-weight model families, grows with scale, and can be moved by narrow finetuning that never touches user turns. We claim that this is evidence that post-training does not merely install a shallow assistant persona, but instead generalises beyond just the local assistant turn, into the model's representation of the user.

This paper develops a sparsity-promoting memory regressor extension (MRE) adaptation law with directional forgetting for nonlinear control-affine systems with linearly parameterized uncertainty. The objective is to use directional forgetting to selectively discount obsolete information and leverage $\ell_1$ regularization to promote sparsity of the parameter estimates. While $\ell_1$ regularization has been applied to the system identification problem in an offline setting, a contribution of this paper is to develop a recursive least squares update law to implement $\ell_1$ regularization in online adaptive control. In particular, we show that $\ell_1$-regularized recursive least squares is realized via a sliding mode update law. A nonsmooth Lyapunov-based stability analysis is then used to show that the tracking and parameter estimation errors are ultimately bounded under a subspace excitation condition. Simulation results on a Van der Pol oscillator demonstrate the ability of the developed sparsity-promoting MRE controller to recover sparse dynamics while maintaining stable tracking.

Density-Ratio Rescoring (DRR) augments a classifier trained at the original class prior with a survey-raking dual score. Raking reweights the majority sample to match minority feature moments within a tolerance. DRR marginally standardizes the dual and base scores and combines them with a fixed weight of one half, using the fitted dual directly for prediction without resampling or refitting the base classifier. Under exact population matching and a correctly specified log-linear tilt model, the dual equals the log density ratio up to an additive constant. A class-separation analysis characterizes the signal strength and correlation conditions under which fusion improves separation under common within-class covariance. On 24 tabular benchmarks, evaluated over 30 trials and five base learners, DRR at the D=128 random-feature setting improves average precision over the standardized base on every dataset, with a mean gain of 0.034. It exceeds the shared-dual raking-and-relabeling resampler on 22 of 24 datasets, with a mean gain of $0.092$, and on all eight one-versus-rest tasks of a shared gene-expression cohort. These results demonstrate the effectiveness of using raking duals as reusable scores for improving rare-class ranking while retaining classifiers trained at the original prior.

We study whether LLMs can produce executable workflow artifacts that remain consistent across graph structure, tool implementation, schema bindings, and runtime wiring. In this setting, correctness depends on cross-layer consistency: a workflow may be structurally plausible, yet still fail because tool implementations, schema bindings, or runtime execution do not align. Existing benchmarks largely evaluate these capabilities in isolation or rely on trajectory-level proxies, leaving open whether generated workflow artifacts execute end-to-end. We introduce \textbf{MCPGen}, an executable benchmark for Model Context Protocol (MCP) workflow development. MCPGen contains 100 self-contained MCP projects across 16 application domains and evaluates three diagnostic tasks: workflow reconstruction, tool creation, and backward-compatible workflow extension. We evaluate 11 representative LLMs in a single-turn foundation-model setting, assessing generated artifacts through static analysis, unit and integration tests, and process-isolated end-to-end execution. Models reach 88.5\% on workflow reconstruction, but no model exceeds 57\% end-to-end execution success. Per-tool unit-test pass rates reach 63.8\%, while project-level integration success does not exceed 45\%, suggesting that integration remains a major bottleneck even when isolated tool tests pass.

Pretrained EEG foundation models are increasingly proposed as general-purpose encoders for brain-computer interfaces, yet recent benchmarks disagree about when their representations transfer to downstream tasks. We audit LaBraM and CBraMod on motor imagery under a validation-locked protocol in which preprocessing, architecture, optimization, freeze depth, checkpoint, temperature, and method selection are determined using training-session data only. On four-class BCI Competition IV-2a, every supervised comparator evaluated here outperforms every foundation-model configuration, including validation-selected fine-tuning. We then examine a key confound: foundation models and task-specific decoders are normally evaluated using different input pipelines. Retraining three supervised architectures on the broadband arrays consumed by the foundation models produces matched-input accuracy differences of opposite sign across architectures: broadband input improves ATCNet by 0.078 accuracy while reducing EEG Conformer accuracy by 0.088. None of the three individual matched-input terms is significant after multiple-comparison correction at n = 9, so we treat the sign variation descriptively rather than as a formal architecture-by-pipeline interaction. These observed sign differences suggest that a single comparator may not provide an architecture-invariant decomposition of a pretrained-versus-supervised performance gap. The four-class deficit also does not reproduce uniformly across motor-imagery datasets: on two-class BNCI2014-004 we cannot detect the same separation between fine-tuned CBraMod and the supervised comparators. Finally, validation-fitted temperature scaling returns foundation-model calibration error to the supervised range despite substantially lower four-class accuracy.

Automated brain tumor segmentation supports diagnosis, treatment planning, and monitoring of disease progression, but building models that generalize across heterogeneous tumors and limited annotated data remains difficult. We present VGG16-MCA UNet, a hybrid architecture pairing an ImageNet-pretrained VGG16 encoder with a decoder in which a Multi-Channel Attention (MCA) module recalibrates features after each skip-connection fusion, trained with the Focal Tversky loss to counter severe foreground-background imbalance. We evaluate the model as a 2D, FLAIR-only, whole-tumor segmenter on tumor-positive slices from two public datasets: the BraTS 2020 benchmark and the LGG MRI Segmentation dataset. Using 5-fold cross-validation and a single network formed by averaging the weights of the five fold models, the method attains an aggregate pixel-level Dice (F1) of 95.10% on our held-out BraTS 2020 split and 88.32% on LGG. These scores are computed over all test pixels pooled into a single confusion matrix rather than averaged per case, and are therefore not directly comparable to the per-case mean Dice used in the BraTS challenge protocol. All partitions were drawn over individual slices rather than over patients, so every patient contributes slices to both training and test; the figures above therefore measure interpolation within known patients and should be read as an upper bound rather than as generalization to new ones. The model segments a 256x256 slice in 66.32 ms on a single 6 GB NVIDIA RTX 2060, approximately 8 ms more than an equivalent VGG16-UNet without MCA. We release the split records and report the protocol in full, with the aim of providing a precisely specified and reproducible 2D FLAIR baseline.

Simulation-based evaluation provides a scalable and repeatable alternative to real-world evaluation of vision-language-action (VLA) policies. However, reconstruction errors can cause simulated policy performance to diverge from real-world performance, motivating the need to assess reconstructed environments for downstream VLA policy evaluation. We present ReVeal, a real-to-sim assessment framework combining workspace reconstruction, reconstruction-level assessment, and matched closed-loop policy evaluation. Novel-View Mesh Fidelity (NVMF) and Annotated Planar Geometry Fidelity (APGF) assess observation and planar geometric fidelity, respectively. We also develop PGSR-D, a reconstruction pipeline incorporating monocular depth supervision to improve geometry where multi-view visual cues are limited. Across 8 assessment scenes, NVMF and APGF consistently distinguish the fidelity of 2DGS, PGSR, and PGSR-D. Matched evaluations of GR00T, SmolVLA, and pi0.5 across 8 humanoid manipulation tasks show consistent ordering between reconstruction fidelity and real-sim performance agreement across pipelines. Further analysis of the evaluation workspaces shows that higher fidelity is associated with stronger real-sim agreement.

Control barrier functions (CBFs) have become one of the most popular tools for encoding and enforcing state constraints in safety-critical robotics. Standard CBF approaches are inherently myopic in nature as they enforce safety only at the current time step. Consequently, the system can be driven toward the boundary of the safe set where no feasible safe control exists at a future timestep. Model predictive control (MPC) based approaches address this by enforcing state constraints over a receding horizon. However, such approaches generally require the model to be known for solving a constrained optimization problem at every step, which is computationally expensive for real-time deployment. We propose BarrierFormer, a barrier-supervised transformer framework that addresses these limitations by encoding rollout-level CBF constraints in learning a model-free safe policy. A causal transformer encodes observation-action history, autoregressively generates a predictive rollout through the dynamics head to replace the model, and provides a residual correction to a nominal controller through the action head to replace the online computation. A barrier critic operating on local observations evaluates CBF constraint violations along this rollout, and a safety teacher computes barrier-consistent actions satisfying these constraints as direct supervision targets for the learned control policy. During inference, the policy maps observation-action history to control actions without any online optimization or model knowledge, enabling real-time model-free predictive safety enforcement. Evaluations across linear and nonlinear, 2D and 3D dynamical systems for safe goal-directed navigation demonstrate that BarrierFormer outperforms existing reinforcement learning (RL)-based, diffusion-based, MPC-based, and transformer-based approaches in safety rate and inference latency.

Agentic AI extends LLM security beyond generated content to persistent state, autonomous actions, tool use, and interactions with humans and other agents. Existing threat classifications often emphasize individual dimensions, obscuring connections among entry points, affected components, and security consequences. The known threat landscape also differs from the coverage demonstrated by empirical research. Through a structured review of 66 studies published from 2022 to 2026, we introduce T={S, B, P, A}, a cross-dimensional representation linking affected functional or system surfaces {S}, interaction or trust boundaries {B}, violated security properties {P}, and empirically examined architectures {A}. We analyze 22 artifact-backed red-teaming studies and 11 representative security benchmarks to characterize empirical coverage and evaluation maturity. Within the selected studies, evidence concentrates on prompt/reasoning, memory, and tool-mediated attacks, predominantly in single-agent settings. Persistent, Human--Agent, complex multi-agent, systemic, and long-horizon threats receive less coverage. These findings describe the selected corpus rather than establish gaps across all empirical research. Heterogeneous metrics, limited adaptive defense evaluation, architectural imbalance, and incomplete execution-state capture further constrain comparison and reproducibility. We derive 13 open research questions to guide more systematic, architecture-aware, and reproducible security evaluation of agentic AI.

Automated kernel vulnerability reproduction is essential for bug triage, patch validation, and regression testing, but still lacks an effective and efficient solution. The core challenge is twofold: a reproducer must first recover the trigger scaffold needed to reach the vulnerable state and determine the precise concrete values that actually trigger the bug. Existing directed fuzzing approaches are ineffective at recovering the necessary trigger scaffold, while LLM-only generation is brittle because it struggles with concrete-value discovery and runtime nondeterminism. We design SyzHarness, a framework that combines LLM reasoning with coverage-guided fuzzing for patch-based Linux kernel vulnerability reproduction. Given a patch, SyzHarness uses an LLM agent grounded by code navigation tools to synthesize a parameterized fuzzing harness that fixes the prerequisite setup logic while exposing only uncertain, bug-critical input parameters to be mutated by Syzkaller. SyzHarness then translates this harness into a Syzkaller compatible interface and iteratively refines it using hierarchical reachability feedback. We evaluate SyzHarness on multiple datasets of triggerable real-world Linux kernel vulnerabilities. On 100 KernelCTF cases, SyzHarness achieves a 78% bug reproduction success rate. On the SyzDirect benchmark, SyzHarness achieves a 73% bug reproduction success rate, substantially outperforming prior directed greybox fuzzing. On 50 recent, known-triggerable syzbot bugs fixed after March 2026, SyzHarness reproduces 40/50 (80%) using only the fix commits as input.

Contact-rich manipulation requires estimating forces, slip and contact geometry that can remain ambiguous in scene images. Optical tactile sensors provide both visual observations of the contact surface and mechanical measurements, yet learning from these signals raises two challenges: representing contact beyond appearance and transferring its benefits to a policy that does not require fingertip observations at deployment. We introduce HapticWAM, a world-action model that combines heterogeneous tactile encoding, structured contact prediction and teacher-student distillation. Its teacher encodes gel images together with deformation, shear, distributed forces, resultant wrench and derived contact state into a frozen video backbone. Rather than predicting tactile pixels alone, the model jointly generates actions and a contact package describing future events and mechanics. Anticipatory Contact Coupling uses the previously imagined package to condition attention, preserving a contact-related input when direct tactile observations are unavailable. Haptic-Imagination Distillation transfers both contact futures and action predictions to a student that retains the generative contact head but removes its fingertip input branches. On a real-world setup, across three contact-rich pick-and-place tasks, HapticWAM Student achieves a 77% per-task mean success rate (41 of 50 starts, 82% pooled), reaching 95% on one of the tasks, outperforming the evaluated teacher and baseline configurations.

Robot foundation models learn manipulation from large demonstration corpora, but surgery is missing from those corpora: across the 780-hour Open-H surgical collection, one dataset carries synchronized force and none covers an aesthetic procedure. Liposuction is the hard case, because the instrument works under the skin and the surgeon operates by feel and by judgment. Humynex Robotics builds curated expert datasets for this kind of procedure. HumynexSurg-1 is the first release: a master liposuction surgeon performing on porcine abdominal tissue while narrating every decision, recorded with synchronized suction pressure, six-axis hand force/torque, top-down RGB-D video, side video and a lavalier microphone -- 14 episodes, 42,738 frames, 35.6 minutes, 356 utterances of which 95% compile into a liposuction-specific label schema. The capture follows a patent-pending sensing plan organized around the quantities a policy needs, so a channel captured today by a model can be upgraded to a sensor tomorrow without changing the data format. This release captures the instrument motion as a tool-hand track in the side video and provides the force channel as state; the funded capture adds a measured 6-DoF handle pose, a validated force channel, ultrasound imaging of the fat layer, and palpation sensing. As a proof of concept, NVIDIA Isaac GR00T N1.7 fine-tunes on the dataset with no custom code in under an hour per run and learns the recorded sessions; scaling probes on the same episodes show where further gains come from: every new session lowers the error on an unseen session. The dataset, its label schema, its quality-assurance reports and its evaluation protocol are the product; the next capture, many short sessions across fat regions with the sensors named here, is what the probes point to.

Joint Embedding Predictive Architectures (JEPAs) are a promising paradigm for learning task-agnostic latent world models without visual reconstruction. However, standard JEPA training exhibits a strong inductive bias towards slow features, causing feature suppression and the collapse of latent representation. While inverse dynamics provides temporal anti-collapse, it relies on action labels and offers little incentive to embed general, unlabeled dynamics. We introduce Difference Image and Single image embedding Regularization (DISReg), a novel regularizer that builds on an inverse-dynamics-style module that predicts temporal difference image embeddings without any pixel reconstruction loss, encouraging balanced static and dynamic feature learning. DISReg consists of a static term that shapes the distribution of the image embedding and encourages slow features, and a dynamic term, which, unlike direct regularization on the embedding, imposes no constraint on the image embedding's shape or distribution and instead only incentivizes that dynamic features be present. By integrating this regularizer into a standard JEPA, we establish our new architecture, MotionJEPA. Latent probing demonstrates that MotionJEPA produces more complete representations than other methods, and our trajectory analysis shows it maintains geometrically simple latent embeddings with low curvature. We further show that MotionJEPA improves downstream planning success under static-background distractors across four environments.

Temporal Information Retrieval (TIR) has been increasingly critical given the rise of Retrieval-Augmented Generation (RAG). Since temporally mismatched evidence can be highly misleading, TIR aims to retrieve documents that are both semantically and temporally relevant to a query. Two TIR paradigms have emerged - temporal retrievers and temporal re-rankers - differing in how temporal relevance is modeled. While these paradigms provide complementary strengths, our analysis reveals that each alone falls short of robust TIR: temporal retrievers provide flexible query understanding via learned representations, but often fail to explicitly account for temporal constraints; temporal re-rankers can enforce such constraints more explicitly, but often rely on predefined re-ranking rules. To address this, we propose Q-TIE, a re-ranking framework based on learned Temporal Intent Extraction (TIE). By introducing a TIE model that maps each query's temporal constraint into a unified interval representation (i.e., $\langle t_{start}, t_{end} \rangle$), Q-TIE generalizes beyond predefined rules via model-based learning while explicitly modeling temporal constraints as a separate signal - jointly achieving what each paradigm typically trades off. Experiments demonstrate that Q-TIE consistently outperforms existing TIR methods with stronger generalizability across temporal query types, and provides a lightweight yet effective add-on for temporally-aware RAG pipelines. Code: https://github.com/ssoy0701/Q-TIE.

The rapid growth of resident space objects is increasing the complexity of space situational awareness sensor tasking, challenging classical optimization methods as they allocate finite, heterogeneous, and distributed sensing resources across ever-larger catalogues. Existing deep reinforcement learning approaches show promise in reduced settings, but fixed-dimensional state and action representations limit their ability to scale to large, dynamic catalogues and distributed sensing networks. We introduce VISTA (Variable-Entity Intelligent Sensor Tasking Architecture), a scalable deep reinforcement learning architecture for persistent uncertainty-driven catalogue maintenance across variable object populations and sensor configurations. VISTA combines physics- and mission-informed top-K retrieval with entity-centric attention, recurrent memory, and pointer-based action decoding, thereby keeping each agent's observation and action spaces independent of catalogue size. We evaluate VISTA across different scenarios, from fixed-size single-sensor benchmarks to large-scale space-based tasking and heterogeneous cooperative sensing. With 30 orbiting targets, VISTA recovers the catalogue 31.2% faster than the fixed-dimensional recurrent baseline. In the large-scale regime, VISTA reduces five-hour uncertainty by 97.5% relative to the strongest classical reference and by 99.3% relative to the recurrent learner. Zero-shot tests up to 20,000 objects reveal near-linear relations between sensing capacity, catalogue size, and recovery horizon. Learned policies also exhibit sensor modality adaptation and generalization to population and initial-uncertainty shifts. Together, these results demonstrate that VISTA provides a scalable framework for adaptive space situational awareness sensor tasking across large, distributed networks of heterogeneous ground- and space-based sensors.

Manipulation policies must know which objects matter and where they are, yet the pretrained backbones that current robot foundation models build on, from language in vision-language-action models (VLAs) to video generation in world-action models (WAMs), do not directly require this metric grounding, leaving it to be learned implicitly from robot demonstrations. We propose Grounded Action Models (GAMs), a new paradigm of robot foundation models built with 3D grounding. GAM can be conditioned using language, points, or box prompts, which are first transformed into a shared object-centric representation of the selected objects. This representation captures target-focused visual features and metric object geometry, which is mixed with robot state history through a multi-stream transformer to predict action chunks. Although GAMs can be run autonomously, they can also serve as a low-level controller that a high-level planner controls using its various input modalities, allowing for long-horizon and memory-dependent manipulation. On RoboTwin 2.0, GAM achieves an average success rate of 55.3% across 50 tasks (vs. 52.0% for Spatial Forcing), including 47.6% under scene randomization (vs. 30.4% for Abot-M0), with its action policy trained only on clean-scene demonstrations. On LIBERO-PRO, it achieves a state-of-the-art average success rate of 61% (vs. 53% for $π_{0.5}$) across 16 perturbation settings, with the largest gains when targets are relocated or newly designated. On two real robots, GAM retains 17/20 successes under visual shift on a bimanual YAM versus 4/20 for $π_{0.5}$, while its composition with a Molmo2 planner on a Franka achieves 64.7% ID and 49.8% OOD step completion on long-horizon and memory-dependent tasks.

Constrained optimization is central to many engineering systems in which decisions must satisfy strict safety and operational requirements, especially in real-time settings with limited computational budgets. In such scenarios, optimization algorithms are often terminated before full convergence, making *anytime feasibility* essential for safe deployment. Existing methods that guarantee feasibility at every iterate typically rely on exact gradient information, an assumption that is often violated in practice due to measurement noise, stochastic approximations, or model mismatch. We develop an anytime-feasible first-order method for nonlinear constrained optimization under norm-bounded errors in the objective and constraint gradients. The method computes a robust search direction by solving a second-order cone program and selects a step size through safeguarded backtracking. Assuming exact function evaluations and a strictly feasible initialization, the method preserves strict feasibility and guarantees sufficient objective decrease whenever the computed search direction is nonzero. We establish a uniform positive lower bound on the accepted step sizes, an O(1/K) bound on the average squared search direction norm, and convergence of the search directions to zero. We also show that a zero search direction at a strictly feasible point certifies approximate first-order stationarity. We validate the proposed method on a multi-agent navigation task in cluttered environments and show that it maintains collision-free trajectories despite noisy gradient information.

Long-horizon robotic search must resolve natural language against heterogeneous, incomplete, and often ambiguous evidence: textual information, prior maps, and observations arriving over time. The core challenge is to contextualize these streams and decide where to gather evidence before selecting a target. We present WORLDS: World-state Observation and Reasoning for Language-guided Discovery and Search, a framework that grounds reasoning in a persistent graph initialized from geospatial priors and updated by perception. Parallel Reasoners maintain competing candidate interpretations and request evidence to distinguish between them. We collect and process the requested observations with a multimodal Examiner, after which a Judge selects a grounded target or requests another pass. WORLDS achieves 51.8% navigation success across all 5,311 CityNav test episodes, the highest reported success rate, exceeding the previous published best by 15.7 percentage points under an OSM-only, high-resolution orthographic protocol. On 1,000 shared episodes, it achieves 50.0% versus 27.9% for the strongest adapted baseline using the same model, prior, sensing stack, and movement budget. Observation-based verification by the Examiner contributes 5.9 points of this success, and at a reduced reasoning-effort setting WORLDS still exceeds the adapted GeoNav baseline by 18.8 points while generating fewer tokens. We also demonstrate WORLDS on a quadrotor, which flies the generated sensing waypoints and grounds three language targets, including a vehicle absent from the map, from its onboard imagery.

Uniform discrete flow permits repeated updates at every generation position. While continued revision supports correction of wrong tokens, it also exposes correct intermediate predictions to later errors. An experiment on Sudoku puzzles shows that 9.4% of generated cells are correct at an intermediate step but incorrect in the final output. We introduce generation order into uniform discrete flow through selective absorption, which fixes chosen predictions while preserving the uniform-flow velocity at active positions. To prevent absorbing incorrect predictions, we propose Low-Entropy Discrete Flow (LEDFlow), a training-free sampler that adaptively orders absorption by local entropy. By decomposing absorption error into joint dependence and conditional prediction terms, we show that selecting the lowest-entropy positions under a fixed absorption budget minimizes an upper bound on the conditional term. We further support the choice of local entropy by showing that the decision-error bound of global lookahead grows with the lookahead window under an imperfect denoiser. Across reasoning benchmarks, LEDFlow attains 0.845 Nikoli Sudoku solve accuracy, with the largest gains on strongly constrained tasks. On text-to-image generation it attains the best overall score, and on multimodal understanding it improves over the native sampler on all six benchmarks, at an inference cost comparable to standard flow sampling.

Memory systems for coding agents must decide, when a repository changes, which of their stored claims have become false. Content anchoring invalidates a claim whenever the artifact it came from changes, which fires constantly. Semantic-equivalence classification asks whether a diff preserves behavior, a question about the diff rather than about any stored claim. We show the second signal fails for a reason unrelated to model capability: asked whether a commit preserves behavior, five models spanning a 40x price range fire on 59-72% of real commits and reach precisions of only 0.291 to 0.329 against a 0.25 base rate. Asked instead whether one specific claim still holds, the same models on the same diffs reach 0.705 to 0.974. A control that hands the behavior-preservation judge the claim text, changing only the question, moves precision by 0.010 and 0.016; changing the question moves it by 0.49 and 0.65. We also compare against pytest-testmon, a deployed regression-test selector with coverage-derived dependency data: it reaches 0.868 recall at 0.415 precision, so near-complete knowledge of what a change can reach does not identify what it falsifies. Ground truth is execution, not annotation: a claim is a test function passing at commit t, and it has flipped if that same assertion text fails at t+1. Building this required an observation we did not find in prior work. On a CI-gated mainline a commit that leaves a pre-existing test failing cannot merge, so the naive construction has an empty positive class by design. We report 10,369 claims with 184 execution-verified flips mined from 23 Python libraries, splits held out by repository, a post-knowledge-cutoff split, a shuffled-diff null, a paraphrase control, and a leave-one-repository-out analysis over 17 repositories.

Automated grading of conjunctival photographs could reduce the cost and variability of trachoma prevalence surveys, but the relative value of modern pretrained visual representations, lightweight feature adaptation, and training-objective design has not been established under a common protocol. This study presents a controlled evaluation for binary classification of Trachomatous Inflammation-Follicular (TF) versus Normal using 1,546 images from the public UCSF/Lietman collection. Images are processed using the OPTED pipeline for zero-shot tarsal-conjunctiva segmentation, alignment, cropping, and standardization. We first compare six pretrained backbones using a common classification pipeline and then evaluate four lightweight adaptation mechanisms on DINOv2 ViT-B/14. Under stratified five-fold cross-validation, DINOv2 with Efficient Channel Attention (ECA) and focal-plus-center loss achieved 91.66 +/- 0.97% accuracy, 90.69 +/- 1.10% macro-F1, and 96.06 +/- 0.71% AUC. ECA introduces only five learnable parameters while matching the performance of substantially larger alternatives. Objective ablation further showed that ECA did not consistently improve plain DINOv2 across loss functions; the lowest-variance 91.66% accuracy was obtained with cross-entropy plus center loss. Overall, the fine-tuned DINOv2 representation provided most of the predictive performance, while ECA offered a highly parameter-efficient refinement whose effect depended on the training objective. The resulting workflow provides a reproducible benchmark for active trachoma image classification.

We present a comprehensive benchmark of Federated Learning (FL) for multilingual Automatic Speech Recognition (ASR), evaluating four Speech-LLM architectures on the Multilingual LibriSpeech dataset. We compare FedAvg and FedProx across frozen and unfrozen encoder configurations, demonstrating that optimized learning rates are critical for performance. Specifically, independently tuning the learning rates for the speech encoder, connector, and decoder yields the lowest error rates, with full three-component adaptation (LoRA for encoder and decoder, full training for the connector) producing the best FL results. We observe that FedProx efficacy is architecture-dependent, providing notable advantages in multilingual pre-trained architectures (e.g., EuroLLM over TinyLlama when keeping the encoder fixed); this indicates that LLM backbone capacity plays a key role in mediating resilience to heterogeneous data distributions. These findings offer concrete design guidance for deploying multilingual Speech-LLMs in privacy-sensitive, distributed environments.

While test-time scaling enhances Large Language Model (LLM) agents in long-horizon software engineering (SWE), sparse binary rewards (Pass/Fail) create a severe credit assignment crisis and waste failed exploratory trajectories. Current trajectory optimization and scaling methods are costly and structurally limited, relying on heuristic state reuse without causal diagnosis or delayed scalar scoring without actionable online guidance. We propose FLARE (Full-Lifecycle Alignment and Reward Engine), a novel dense supervision paradigm driven by a lightweight Generative Reward Model (GRM). First, RADAR, an offline causal-aware diagnostic framework, extracts high-fidelity, hindsight-free supervision through causal-chain backtracking to distill a GRM providing real-time, step-level risk feedback. Second, FLARE uses this GRM to continuously optimize the agent across its entire lifecycle. During inference, FLARE acts as an Active Scaffold, autonomously intercepting high-risk generation steps for localized breakpoint re-execution, drastically reducing compute overhead. During post-training, the GRM's structured signals serve as process-supervised reranking scores for Supervised Fine-Tuning (SFT) and step-level dense rewards for Reinforcement Learning (RL), mitigating policy collapse in sparse environments. Extensive evaluations show that FLARE establishes a new Pareto frontier across the agent lifecycle: FLARE (N=1) outperforms Global Rollout (N=5) with a 5x reduction in token consumption. Extending FLARE to training overcomes the sparse reward problem in long-horizon interactive tasks, delivering relative performance gains of 19.13% in SFT through process-aware data curation and a consistent 9.19% improvement in RL.

Many knowledge-work benchmarks are constructed around individual tasks, with the context needed for each task selected together with or after the task has been specified. This design measures performance on workplace-like tasks in an environment assembled for the task. When task specification guides which context is selected, the evaluation can encode task information into the environment and pre-complete part of the information-localization work that workplace performance normally requires. We introduce WorkWorlds, an evaluation infrastructure that separates organizational state from task specification. A world first fixes a revision, date, and employee seat and materializes the organizational state that employee can access; tasks are introduced only afterward. We implement WorkWorlds in a primary synthetic pharmaceutical company with 8 measured tasks across 6 employee seats, and construct additional organizational worlds. Across 192 matched evaluations, task-level curation increased evidence access by 17.6 percentage points, from 72.8% to 90.4%, and criterion pass by 8.7 points, from 68.0% to 76.7%, while pass conditional on evidence access remained nearly unchanged; most of the measured difference occurred before the agent reached sufficient evidence.

High-precision connector insertion remains challenging for robotic systems due to tight mechanical tolerances, partial observability during contact, and multimodal uncertainty arising from occlusion and contact ambiguity. Successful insertion requires closed-loop contact guidance that continuously integrates global alignment cues with local contact feedback to produce stable corrective actions under interaction. In this work, we present ContactDP (Contact-Guided Diffusion Policy for Tight Insertion Tasks), a multimodal diffusion-policy framework for contact-rich insertion. ContactDP jointly integrates wrist RGB observations, fingertip tactile sensing, and wrist-mounted force-torque measurements to infer contact state and generate temporally consistent corrective motions during insertion. To ensure stable execution under contact, the learned policy operates together with a hybrid position-force controller that provides compliant low-level interaction. We evaluate our approach on a suite of industrial-grade connector insertion tasks with varying connector geometries, grasp conditions, and initial misalignment. Across all tasks, ContactDP significantly outperforms vision-only diffusion policies for performance, reliability and generalization.

Human motion is shaped by both external acoustic events and behavioral intent: spatial audio conveys environmental cues that elicit or guide a response, while text specifies the desired action and how it should be performed. In this paper, we study the novel task of human motion synthesis jointly conditioned on spatial audio and natural language, a problem that has been largely overlooked in previous research. To support this task, We introduce STAM, a dataset of motion sequences paired with spatial audio and detailed textual annotations whose rich vocabulary affords precise and nuanced specification of human motions. We further introduce MoSAT, a latent flow-matching framework for full-body motion generation jointly conditioned on natural-language intent and directional spatial-audio cues through hierarchical cross-attention before generating motion. Such a hierarchical design enhances temporally coherent and semantically aligned motion sequences. We also develop tri-modal evaluators for comprehensive evaluation on this novel task. Extensive experiments show that MoSAT achieves the SOTA performance by leveraging spatial audio's intrinsic motion-shaping properties alongside textual semantics, enabling precise and diverse motion in various scenarios.

Every node in a multi-agent large language model (LLM) workflow retrieves context from memory and injects it into its prompt, where those injected tokens are billed as input tokens at the same per-token price as the system prompt and the user query. Production observability tools report total token cost but do not separate the tokens a node generates from the tokens it is handed, so this component of the bill is invisible to the teams paying it. We introduce the Total Cost of Agency (TCA), a decomposition of multi-agent workflow cost into base prompt, inference, memory injection, miss penalty and context-accumulation components, and an exact attribution method: a two-pass, non-billable token count that measures injected tokens directly rather than estimating them from word-count proxies. On a 200-task enterprise benchmark executed against real model APIs, memory injection accounts for 13.6 percent of the variable cost a compile-time optimizer can act on, about 12 percent of the full billed cost, and its share rises from a structural zero at workflow depth one to 27.6 percent at depth six. Injected tokens grow linearly with depth over the measured range (R^2 = 0.9974, depths two through six); a quadratic fit yields a negative leading coefficient, so the data do not exhibit convex growth at these depths. We show the component is controllable at fixed model tier: reducing the retrieval window capacity from 32 to 2 entries lowers injected tokens by 28.7 percent with an accuracy change within seed-level variation. We report in full that our graph-rewriting transforms are approximately cost-neutral in isolation, that two of the five decomposition terms are zero by construction in this harness, and that total workflow cost is dominated by model tier assignment, which we hold fixed and treat as prior work. Prompt caching is not evaluated; all figures are for the uncached case.

Robotic bin packing requires long-horizon sequential decision-making, as each object placement affects the available space for subsequent packing. Existing methods primarily rely on hand-crafted geometric heuristics that optimize predefined objectives or reinforcement learning policies learned through trial and error over predefined training configurations. Despite recent advances in multimodal large language models (MLLMs) for this task, their potential for closed-loop sequential decisions across heterogeneous packing configurations remains underexplored. To address this gap, we introduce PackLab, a comprehensive framework for developing, training, and evaluating MLLMs for closed-loop robotic bin packing. PackLab-Suite provides a physics-based simulation platform for scalable generation of diverse training packing trajectories and evaluation of their physical outcomes. PackLab-VLM is a packing-specialized MLLM that understands the evolving object and container states to jointly select objects and predict placements in a closed-loop manner. PackLab-Bench provides standardized packing scenarios at multiple difficulty levels for systematic evaluation. Extensive experiments demonstrate that, on average, PackLab-VLM outperforms conventional packing heuristics, traditional reinforcement learning methods, and general-purpose MLLMs across object sets and container configurations, highlighting the potential of MLLMs for long-horizon robotic packing. The code, model, dataset, and benchmark are available at https://github.com/Correr-Zhou/PackLab .

Many real-world decisions require prioritizing high-risk cases, such as clinicians prioritizing high-risk patients before lower-risk ones. Falling rule lists (FRLs), which are ordered if--then rules with monotonically decreasing risks, provide an interpretable framework for such tasks; however, their single-path structure yields a highly restricted model class. We introduce falling trees, a new family of interpretable models that enforces the same monotonic risk constraint while permitting tree-structured branching. We present GRAVITree, a novel dynamic-programming-with-bounds algorithm for learning the Rashomon set of falling trees under depth and branching constraints. Our formulation can interpolate between rule lists and full decision trees, enabling user-desired model expressivity. In a new clinical dataset and in many public classification benchmarks, falling trees match or outperform FRLs and other interpretable baselines, often producing more sparse decisions for high-risk instances. Our results show that falling trees strike a practical balance between interpretability, expressiveness, and risk prioritization for high-stakes settings.

Decentralised federated learning replaces server aggregation with peer-to-peer model exchange, making collaborator selection a local decision under uncertainty. Fixed probe budgets waste effort on easy choices yet fall short when peers are hard to distinguish. We propose SAGE (Sequential Anchor-Gated Exchange), an optimal-stopping peer selector under a one-model-bearing-exchange budget. A receiver scores candidate neighbours on receiver-owned anchor evidence and selects once an advantage is certified. It continues probing only while further evidence repays its cost, and otherwise falls back to random gossip. We show that the stopping problem admits an optimal rule attained at a finite stage, and that the anchor schedule is order-optimal in the peer-risk gap and the confidence level. We further show that the selector never returns a peer worse than random gossip with high probability, and prove that no such guarantee holds for selectors that commit without a certificate. A separability threshold follows, below which no probing budget improves on gossip. Experiments span two image benchmarks, two graph families and three heterogeneity levels. Selectors that always act on their evidence lose to gossip in every configuration tested. SAGE-OS matches gossip on 75.5% less evidence than a fixed budget, at half the communication overhead of two published selectors. The operative decision is not which peer to rank first, but whether the evidence justifies ranking at all.

The disambiguation of semantically similar statutory text across jurisdictions is a retrieval problem that existing methods do not solve. This inter-context conflict can steer generative models toward confidently produced answers grounded in topically relevant but jurisdictionally incorrect sources. Tobacco and nicotine regulations vary by US jurisdiction, often sharing similar language, thus, robust reasoning requires identifying which jurisdiction's law governs a given product, not merely retrieving relevant text. Emerging products (e.g., pouches) exploit ambiguous definitions to evade regulation. State-of-the-art (SOTA) document retrieval-augmented generation (RAG) methods struggle to address this inter-context conflict, and thus struggle to connect image attributes (e.g., rich attribute captions) to the set of similar legislation texts. We introduce NicoPRISM (Nicotine Product and Regulation Image-and-Text Surveillance Multimodal), comprising 161,563 images, attribute captions, a knowledge base of product, health, and legislative documents spanning 13 US jurisdictions, and 1,495 validated question-answer pairs across two tasks: policy compliance QA and product knowledge QA. We also propose PRISM-RAG, a multimodal hypergraph RAG framework built over images, captions, and entities without any LLM calls at index time, grounding every query in a product image and routes retrieval through a jurisdiction-aware context assembly mechanism guaranteeing that statutory text from the queried jurisdiction reaches the language model by construction. PRISM-RAG retrieves passages from the correct jurisdiction in 93.9% of policy compliance queries, a 48.6 percentage point advantage over standard RAG (p<0.001), using zero LLM calls at index time and one at query time, and is competitive with or outperforms SOTA RAG frameworks across keyword, semantic, jurisdiction-, and compliance-accuracy metrics.

Root cause analysis at a remote site is slow: evidence is scattered across pod logs, Kubernetes events and cluster-level objects, and many operators cannot send production logs to a hosted model at all. On-premise inference removes the second constraint but raises a question live-cluster benchmarks have not addressed: when one workstation GPU fixes both the model and the context budget, how should evidence be retrieved, and what happens when the runbooks the model consults have been tampered with? We present TriFleetRCA, a pipeline running entirely on one on-premise GPU that collects evidence at one of three scopes (pod, namespace, cluster), ranks it by template de-duplication then BM25, filters runbooks through an ingest guard, and returns a root cause with the evidence lines supporting it. We evaluate on a live Kubernetes cluster into which we inject four faults, so ground truth is known by construction, across 100 analyses with Qwen2.5-14B-Instruct at temperature 0. The hit rate was 0.85, 0.90 and 0.95 at pod, namespace and cluster scope; intervals overlap, but the whole scope effect comes from the one fault whose cause is a cluster-level object, and cluster scope costs 55% more tokens. De-duplication before ranking raised the hit rate from 0.75 to 0.90 at equal token cost. A poisoned runbook telling the model to delete the namespace was rejected by the guard every run; with the guard disabled the model declined to follow it in all 20 analyses, making the guard defence in depth rather than the sole barrier. Separating citation quality from accuracy proved informative: one fault was diagnosed correctly and cited incorrectly every trial, a failure mode accuracy conceals. Median latency was 1.6 s at 2,200 prompt tokens. We release the pipeline, the fault injector and all records.

Few-shot recognition with frozen visual features is especially fragile under domain shift and one-shot supervision, where a single labelled image is an unreliable estimate of its class. We ask how far this fragility can be reduced purely at test time, without retraining the encoder or augmenting the source domain. We present Spectral Transductive Refinement (STR), a training-free transductive inference rule that exploits the geometry of the complete support-query episode. Given frozen embeddings, STR builds a joint k-nearest-neighbour graph, maps the episode into a normalized-Laplacian spectral coordinate system, initializes class representatives from the labelled support, and iteratively refines them using pseudo-labelled queries. We evaluate STR under two protocols. A controlled component study with frozen ResNet-18 features shows that spectral refinement consistently improves over single-prototype spectral initialization across five shifted domains, with the largest gains in the one-shot regime where support estimates are weakest. We then benchmark STR against recent Cross-Domain Few-Shot Learning (CD-FSL) methods using the standard miniImageNet-pretrained ResNet-10 backbone over eight established target domains. Operating entirely at inference time, STR attains the highest 1-shot average among compared methods and remains competitive at 5-shot, rivalling approaches relying on heavy source-domain meta-training augmentations. Because STR is transductive, we report its setting explicitly. Diagnostics attribute its gains to iterative refinement in spectral coordinates rather than added prototype capacity, which remains inactive in our configuration.

We demonstrate a simple and effective approach for realizing extended volumetric optical trapping using a single linearly polarized Gaussian beam propagating through a stratified medium with engineered refractive index gradients. On modifying the gradients, the focal field undergoes axial elongation and develops multiple localized intensity maxima, enabling simultaneous trapping of particles at distinct axial planes. Both on-axis and off-axis trapping configurations are experimentally observed, forming a volumetric particle distribution without the need for complex holographic beam shaping. A rigorous theoretical framework based on the Debye Wolf diffraction formalism combined with Generalized Lorenz-Mie Theory is employed to model the focused field and the resulting optical forces. The analysis, substantiated with numerical simulations, shows that spherical aberration arising from refractive-index discontinuities plays a crucial role in redistributing optical energy, leading to the formation of multiple stable trapping regions. The axial trapping potential decreases monotonically with increasing mismatch, indicating a trade-off between multi-site trapping capability and trapping potential, which is in quantitative agreement with experimental observations. This work establishes refractive index engineering as a powerful tool for tailoring three-dimensional optical force landscapes using minimal optical complexity, thus offering a scalable alternative to conventional holographic optical tweezers towards applications in colloidal assembly, micromanipulation, and biophotonics.

Egocentric human data provide a principled source of supervision for learning dexterous robot manipulation. Unlike prior approaches that often collect such data in constrained or specially constructed environments, we collect in-the-wild egocentric demonstrations in real-world settings, including homes, factories, and pharmacies, etc., where people perform their ordinary tasks while wearing head-mounted cameras. This collection protocol captures diverse workflows and hand-object interactions across long-tailed object and skill distributions, but also yields visually challenging observations due to scene clutter and head-motion-induced viewpoint changes (a mean cumulative rotation of $15.93^{\circ}$/s). To address these issues, we introduce EgoWild2Dex, which transfers in-the-wild ego-human experience to dual-arm robots with dexterous hands by jointly aligning unstable egocentric views and human motions with robot observations and actions, respectively. This work offers three benefits. First, we introduce GeoFormer, a differentiable geometric transformer that warps noisy human observations toward robot observations. Second, we design a human-robot training scheme to bridge the embodiment gap, enabling high task success with limited robot supervision. Third, we release EgoWild, a 538.9-hour in-the-wild egocentric human dataset comprising 179,049 episodes, 125,961 unique task descriptions, and 1,282 object categories. On real robots, EgoWild2Dex achieves an average success rate of 96.7% across three long-horizon bimanual dexterous manipulation tasks and an average object-level zero-shot success rate of 33.3%. The data, models, and code will be released.

Generative world models aim to predict future states conditioned on actions, where action controllability is fundamental for reliable dynamics modeling. While recent efforts leverage simulator-generated data to enhance this capability, existing training pipelines face two fundamental limitations. First, static offline data collection leads to a distribution misalignment between training sets and the model's evolving error patterns, failing to resolve critical long-tail scenarios where dynamics predictions remain unreliable. Second, the standard objective of minimizing observational discrepancy often encourages the model to exploit spurious correlations instead of capturing the underlying action-effect causality. To address these limitations, we propose OnlineWM, an online training framework that continuously improves world modeling through active simulator interaction and causality-aware optimization. OnlineWM introduces two key innovations: (1) Active Online Learning: Instead of using fixed datasets, OnlineWM adaptively queries the simulator for new interaction sequences that target the model's current predictive weaknesses, ensuring high-utility data acquisition. (2) Causality-Aware Fine-Tuning: We propose a counterfactual learning strategy that contrasts the outcomes of different actions from identical states, forcing the model to attribute state transitions to specific actions rather than ambient environmental evolution, thereby grounding its predictions in reliable causal mechanisms. By integrating active data acquisition with causal optimization, OnlineWM establishes a closed-loop refinement process that ensures the model is both robust to diverse scenarios and precise in its causal attribution. Extensive experiments demonstrate that OnlineWM significantly enhances action controllability and generalizes effectively to unseen domains.

Autonomous multi-drone navigation requires fleets to service distributed objectives in cluttered environments under tight computational budgets. Efficient coordination depends on task bundling, where each drone visits multiple objectives along its route. Separate solvers for cost estimation, assignment, and execution incur redundant graph search and produce long, abrupt paths. We propose FlockDiffusion, a learned framework combining a scene graph encoder, an explicit allocation head, an assignment conditioned diffusion transformer, and a closed form trajectory decoder. An autoregressive teacher provides offline supervision for parallel fleet trajectory generation. PyBullet ablations show that bundling increases task completion from 50% to 100%, while our complete teacher further reduces route cost by 8.4% relative to MAGNNET with bundling. In the optimized scalability benchmark, evaluated on 100 scenes per density with ten drones, FlockDiffusion achieves 6.2 to 7.6 times faster inference and approximately 37% shorter routes than the classical pipeline. As nominal task counts increase from 20 to 40, latency rises from 7.8 to 11.1 ms, compared with 48.0 to 75.8 ms for the baseline. In a separate evaluation across five Gazebo environments, FlockDiffusion achieves 100% planner coverage and reduces planned route cost by 15.4% relative to the baseline with bundling. These results demonstrate efficient planning under increasing task density in configurations that are demanding to reproduce with physical drone fleets.

Small open-source large language models (LLMs) in the 0.6B-4B parameter range are increasingly deployed for structured output generation (JSON, function calling, data extraction), yet little is known about how constrained decoding (CD) interacts with model scale in this regime. We benchmark five models from three families across 14 structured-output tasks under three decoding conditions (native, Outlines, XGrammar). We introduce a two-axis evaluation that separates structural correctness (schema validity) from semantic correctness (content accuracy). We find that CD eliminates all structural failures across all models (schema validity: 78.6-92.9% to 100%), but content accuracy reveals a persistent semantic gap that is scale-dependent: type coercion failures are fully CD-rescuable, while instruction-semantic failures (e.g., multi-step function calling) remain CD-resistant. Schema conformance is necessary but not sufficient for semantic correctness; CD's reach ends exactly where schema conformance ends.

Large language models have shown strong performance across a range of legal tasks, but existing benchmarks rarely evaluate the ability to take and defend a legal position, reason under incomplete information, or synthesize multiple statutory provisions. This gap is particularly pronounced for Canadian law, which remains underrepresented in legal NLP. We introduce GRACE (Grounded Reasoning Adversarial Canadian LEgal examples), a dataset of 1,915 question-reasoning-answer instances grounded in Canadian federal legislation. GRACE covers three reasoning modes: adversarial advocacy, uncertainty, and applied reasoning. We develop a pipeline that partitions raw statutory text, generates scenario-based questions and reasoning, and filters examples through model-free citation verification and LLM-based quality auditing. As a proof of concept, we fine-tune CLeAR-4B (Canadian Legal Adversarial Reasoning), a lightweight model for grounded legal reasoning, and evaluate it against the unmodified Qwen3-4B base model in open- and closed-book settings. CLeAR-4B substantially improves agreement with teacher outputs and statutory citation behavior when the relevant act text is provided, while its grounding degrades sharply when the statute is withheld. These results suggest that GRACE can support the development of lightweight legal models that reason more effectively from supplied statutory text.

Frontier benchmarks need tasks that current models cannot solve. But a task that no model solves is not automatically a hard task. The same zero pass rate can come from a real capability gap, but it can also come from missing context, a broken reference solution, infrastructure failure, or a verifier that can be bypassed. In this paper, we study this issue using a frozen Terminal-Bench 3 / Frontier-Bench 0.1 production record with 1,081 pull requests, 639 scored tasks, 28,801 trials, and $105,933 in logged agent spend. We ask what an all-fail task actually certifies. For the 125 tasks with no honest pass, we combine task artifacts, reference-solution runs, empty-solution controls, adversarial trials, trajectories, telemetry, and review records, and apply an ordered validity screen. Only 78 of the 125 tasks survive as certified-unsolved candidates. The remaining tasks include 14 with broken oracles, 8 dominated by infrastructure failures, 4 that are only passable through verifier bypasses, and 21 whose solvability is not certified by the available evidence. Thus, lack of saturation and genuine difficulty are not the same thing. The certified-unsolved label is also narrow: it means that the authored route passed, infrastructure did not dominate, no strict bypass was observed, and all evaluated agents failed. It does not prove intrinsic hardness, verifier completeness, or failure at the intended capability. We further analyze rejected submissions and passing tasks to show that pass rate alone cannot explain why a task is difficult. Overall, our results suggest that frontier benchmarks should report the evidence behind their all-fail tasks before using them as capability claims.

Global vision--language similarities compress an image and a caption into one vector, preserving semantics but not which word corresponds to which region or how those regions are arranged; a model can recognize every word and object yet prefer a compositionally incorrect caption. We argue that a frozen encoder retains this association structure, so the problem is to read it rather than to rebuild it beside the pretrained similarity. We introduce BindCLIP, a pairwise scorer built on one latent object: a balanced token--patch--depth optimal-transport coupling that places both candidate captions and several visual depths in a single plan. Semantic, entity, order, and spatial evidence are read as energies of this state, and exchanging the candidates permutes the plan, making the score exactly antisymmetric. A geometric refinement inside the coupling contracts moves that the candidates and the visual depths do not support. No task label, parser, relation inventory, or detector is used. One checkpoint and one inference path improve the official What'sUp, ARO, and SugarCrepe benchmarks over frozen global CLIP, with the strongest transfer on the relation splits. Controls rule out patch access and caption-length shortcuts, and an inference-time lesion localizes spatial arrangement to the coupling.

Textual-gradient methods automate prompt optimization through natural-language feedback, but their iterative updates can be unstable. We identify two sources of this instability: noisy gradients produced from already-correct examples and over-specialization to hard cases that degrades performance on simpler inputs. We introduce STEVE, a stabilization framework with two coupled mechanisms. Error-Driven Refinement generates gradients only from incorrectly handled examples, concentrating updates on informative failures. Regularized Verification treats every update as provisional and accepts it only when improvement on hard cases does not cause unacceptable regression on a preservation set. Across ten reasoning benchmarks, three evaluator/optimizer models, and established prompt-optimization baselines, STEVE reduces degradation and produces more robust prompts. Additional evaluations with gpt-5.4-mini/gpt-5.4 on symbolic reasoning, GSM8K-Platinum, and DS-1000 show that these gains persist with newer models and larger test sets. STEVE therefore provides a practical way to improve the stability and effectiveness of textual-gradient prompt optimization.

Multimodal large language models (MLLMs) incur substantial computational overhead due to the reliance on hundreds of visual tokens to represent images. While token pruning has emerged as a promising approach to reduce the inference cost of MLLMs, existing methods typically reassign position embeddings to the retained tokens using either sparse or continuous position embeddings, each introducing distinct limitations. Sparse position embeddings tend to decrease the attention value allocated to visual tokens, thereby degrading the perception capability of MLLMs, whereas continuous position embeddings disrupt the original spatial correspondence of visual tokens, leading to weakened grounding capability. To mitigate this issue, we perform layer-wise analysis of the language decoder and observe that intermediate layers play a critical role for maintaining the grounding capability of MLLMs under token pruning. Based on this observation, we propose a layer-aware position embedding strategy, which switches to sparse position embeddings at grounding-sensitive layers while maintaining continuous position embeddings elsewhere. Extensive experiments across representative pruning methods and diverse benchmarks demonstrate that our approach improves the comprehensive multimodal performance of pruned MLLMs compared with standard sparse and continuous position embeddings.

Large language model agents are now privileged principals that take consequential actions: editing code repositories, operating inboxes, completing purchases. Their authority is kernel-grade, but it comes without what classical systems security requires: a trusted mediator interposed on every access. Operating-system vendors are now rebuilding the platform around this de-facto agent kernel, inheriting complete mediation as a design problem. We systematize the security of such systems around a single distinction: a crossing mediated over provenance admits a deterministic check, while one over content semantics does not. A trust-boundary taxonomy locates where mediation must occur and isolates the central mediation gap at two kinds of semantic judgment: distinguishing data from instruction in untrusted input, and an authorized action from an unauthorized one. We argue that this gap leaves an irreducible residual of undetected attacks wherever inputs and actions are not restricted in advance to an enumerated set. The same distinction makes attack-success statistics actionable, placing each number on a spectrum from deployment debt (a sound deterministic mediator left unused) to a structural gap (no such mediator known). We systematize defenses across runtime monitoring, architectural separation, and authorization, and show that current evaluations tend to overstate deployed security through evaluation-validity failures. Finally, we carry that analysis forward beyond the de-facto kernel, to an architecture in which the model itself becomes the arbitration core, and derive the design constraints, open challenges, and research agenda for a security-first AI-native OS.

Recent advances in Physical AI have accelerated the use of foundation models in autonomous systems such as unmanned aerial vehicles (UAVs), which must perceive, reason, plan, and act in dynamic environments. Existing benchmarks assess physical perception, intuitive physics, embodied navigation, and collaborative reasoning, but rarely evaluate the agentic decision-making required for reliable autonomy. We introduce \textit{PhysAI-Bench}, a benchmark for evaluating this capability. It contains 10,178 standardized decision instances automatically extracted from conversational traces of autonomous UAV missions. Each instance preserves mission context, temporal dependencies, physical constraints, Model Context Protocol (MCP) tool calls, Agent-to-Agent (A2A) interactions, sensor observations, and AI-native 6G network conditions, including latency, packet loss, throughput, edge load, and network slicing. We expose only information preceding each decision, preventing future-event leakage and approximating online decision-making. We evaluate 29 foundation models using a two-stage protocol. We select model-specific configurations from 12 combinations of zero-, three-, and five-shot prompting and four temperatures, tested in three runs on a 35-instance, human-verified development set. We then freeze each selected configuration and evaluate it in three runs on a fixed, episode-disjoint set of 500 instances. GPT-5.3 achieves the highest accuracy (52.00%), followed by GPT-5.2 (49.40%) and Grok~4.5 (49.07%). Few-shot prompting generally improves performance, while temperature has limited influence. The results demonstrate that reliable agentic decision-making in Physical AI remains an open challenge. The dataset is available at https://github.com/maferrag/physai-bench

Co-stabilization, i.e., designing a single controller that stabilizes multiple systems, is a fundamental problem in robust and data-driven control. While any stabilizable linear system admits a stabilizing linear static state feedback controller, this equivalence does not extend to co-stabilization. In particular, there exist system collections that cannot be co-stabilized by linear static state feedback but can be co-stabilized using linear dynamic state feedback. In this paper, we study the role of controller memory in co-stabilization. We show that linear dynamic state feedback strictly enlarges the set of co-stabilizable systems compared to static feedback, for both scalar systems and high-dimensional examples. At the same time, we identify structural limitations that cannot be overcome even with dynamic controllers. We also develop a path-integral-based algorithm for computing co-stabilizing controllers for a finite set of systems. Numerical results demonstrate that increasing controller memory enlarges the feasible co-stabilization region. These results highlight controller architecture as a key structural factor in co-stabilization, with potential implications for reducing the sample complexity of learning-based control.

Post-training quantization of speech language models is often summarized with text-output scores and nominal bit widths. Those numbers alone do not establish behavior that depends on information missing from a transcript, or efficiency for a particular runtime. We introduce an evaluation protocol that separately tests lexical output, a transcript-insufficient endpoint, and a measured packed implementation. In a Qwen2-Audio case study, a translation-selected 6-bit allocation improves chrF by 2.36 on a frozen English-to-German replay, with paired 95% bootstrap interval [1.04, 3.62], but loses 3.91 percentage points on speaker-disjoint emotion recognition. At the same 6-bit budget, the uniform structural control reaches higher emotion accuracy than the selected allocation, and the front-layer control is also higher by point estimate on the same frozen set. At 7 bits, chrF improves by 3.28 with interval [2.08, 4.59], the emotion interval against FP16 includes zero, and a same-budget front-layer control still exceeds the selected allocation. A separate matched-budget 4.08-bit study finds roughly 10-point emotion deficits for every tested low-bit allocation and no selected-allocation advantage over frozen controls. Finally, a dequantized average-6-bit simulation retains the FP16 peak memory. This case study identifies a precision-dependent mismatch between lexical output, waveform-dependent behavior, and nominal precision. It does not establish a general failure of low-bit speech models or a deployment benefit for the selected allocation.

Form Field Detection (FFD) is a fundamental component of document understanding systems, enabling applications ranging from large-scale industrial digitization to accessible form interaction for automated analysis. Unlike conventional object detection tasks, FFD is inherently challenging because fields are often defined by layout structure and whitespace rather than visible foreground content. Existing large-scale datasets frequently rely on heuristic annotation pipelines, resulting in noisy and inconsistent labels that hinder reliable evaluation. In this work, we introduce mini-CommonForms, a carefully curated FFD benchmark with consistent, high-quality annotations, and present a detailed evaluation of state-of-the-art detection approaches. The benchmark is designed to support reproducible research in document automation and accessibility-oriented applications. Dataset and code are available at https://github.com/moured/mini-commonforms

Open-world deployment requires humanoid robots to cross highly heterogeneous terrain safely, with perception that simultaneously provides wide coverage, local accuracy, and redundancy against sensor failure. Existing approaches struggle to satisfy all three: one forward depth camera or nearby height sampling covers too little; odometry-corrected elevation maps drift under aggressive motion and miss thin vertical structures; image-level encoding costs grow with camera count. We present UniPoint, a humanoid whole-body locomotion framework built on multi-source point-level sensor fusion. Measurements from a 360° light detection and ranging (LiDAR) sensor and two depth cameras are early-fused into one base-frame point set. Voxelization resamples it to a fixed number of tokens encoded by linear self-attention and proprioception-queried cross-attention, decoupling forward cost from sensor count. The point set retains standing thin barriers; a single-modality failure removes only part of the tokens, so the policy degrades gracefully. A single training run with terrain-aware rewards, perception-degradation injection, and domain randomization produces one policy for all eight terrain types, deployed on an onboard RK3588 without fine-tuning. On a DR02 humanoid, 20 trials at each of nine real-world settings over seven terrain types validate the policy on 70-cm-high platforms, 100-cm gaps, thin barriers, and sparse or narrow footholds; it also generalizes zero-shot outdoors.

Constraint-following music generation asks a score to satisfy several user-specified properties at once, each checkable programmatically (key, meter, length, range, final note, rhythm, motion and form), yet no existing benchmark isolates this capability. We construct MusicConstraintBench, 2,180 items over eight constraint families, on which current models fail once a few constraints are combined. The natural remedy is reinforcement learning with these verifiers as reward, yet we observe that a reward paid only when every property holds leaves most training groups without a learning signal: over the first 50 updates, 0.550 of rollout groups score identically and receive no gradient, even though a failing score typically misses only one requested property. Under the joint criterion, rollouts for a prompt tend to fail together, so a binary reward cannot separate a nearly correct score from a malformed one. We therefore introduce MusicRLVR, which pays graded per-property credit behind a hard validation gate that rejects malformed outputs, plus a joint-satisfaction bonus, requiring no human annotation, learned reward model, or music-domain fine-tuning. On MusicConstraintBench, MusicRLVR lifts Qwen3-4B-Instruct from 0.160 to 0.807 on mixed constraints and leads every zero-shot baseline including Llama-3.1-70B at 0.380. It also generalises to property combinations unseen in training and to out-of-range parameter values, showing that verifiable rewards need not presuppose a target output.

Ethereum decentralized finance (DeFi) provides a public, time-stamped record of transaction-level event streams, but the same public symbols can create strong machine-learning shortcuts. We introduce ETH-TraceBench, a benchmark for evaluating Ethereum DeFi representations under temporal, protocol, pool/infrastructure, and symbolic shift. The raw event universe covers January 2021-December 2025 and contains 1.35 billion transactions with logs and 5.01 billion raw log rows. Model evaluation uses a fixed 911,267-instance supervised sample, training on 2021-2024, selecting models on 2025H1, and testing on 2025H2. Simple models perform strongly on the aggregate temporal test: TraceStats-GB reaches 0.953 macro-F1 and TopicEmitterHashMLP 0.959 on the canonical DEX test set. Performance drops sharply under protocol novelty, with macro-F1 of 0.794, 0.743, and 0.766 for TraceStats-GB, TopicEmitterTrace-SGD, and TopicEmitterHashMLP, while strict unseen-pool scores remain 0.927, 0.897, and 0.935. Uniswap v4 and Ekubo v1, both absent from supervised training, are materially harder than the full test. Jointly masking emitter and topic identity reduces DEX macro-F1 to 0.916 and liquidation macro-F1 to 0.774 for TopicEmitterTrace-SGD. A standard Transformer over log-index-ordered events provides no consistent advantage over a deterministic shuffle of the same events, indicating that high aggregate scores can arise without sophisticated chronological modeling. A natural-prevalence audit estimates 2025H2 DEX prevalence among logged Ethereum transactions at about 22.5%, and a deterministic 400-transaction audit finds complete agreement with task label sources and independently re-queried raw-log counts. ETH-TraceBench therefore treats difficult transfer and controlled-input conditions, rather than a single aggregate score, as the main evaluation target.

LiDAR-based unmanned aerial vehicle (UAV) exploration builds maps by continually selecting where to observe next. However, decisions based on the measured map provide limited foresight into spatial continuations behind occlusions, leaving potentially informative directions unrecognized. We present WOLF, a world-model-guided framework that predicts future observations to enhance autonomous exploration. In the training stage, a recurrent world model learns observation dynamics from exploration trajectories, with recurrent memory retaining the spatial context needed to interpret partial observations across successive views. Building on this context, the model combines observation history with candidate motions during exploration to predict local occupancy and visibility. To guide further sensing, a predictive frontier generation mechanism then aligns and fuses these predictions using confidence, branch agreement, and observation quality to identify promising regions. The resulting predictive frontiers join measured ones to guide geometric viewpoint selection and trajectory generation, while new scans update subsequent predictions. In simulations, our method reduces mean terminal time by 10.9% relative to EPIC in Garage at comparable coverage and increases mean coverage from 42.12% to 98.35% in Tunnel. Real-world experiments further demonstrate onboard deployment of the learned model for online inference during physical flight.

Vision-language-action (VLA) models have achieved strong performance in embodied manipulation, but still lack a clear mechanism to balance behavioral stability with task-semantic sensitivity. We identify two complementary failure modes. Under task-preserving changes, where task semantics remain unchanged but scene appearance varies (e.g., style, illumination, clutter, or paraphrasing), policies often exhibit unnecessary action drift. Conversely, under semantic-breaking changes, where key task semantics such as the target object or constraint are altered, policies frequently fail to produce sufficiently distinct behaviors and instead follow the original trajectory. To address this gap, we propose BAS-VLA, a task-semantic action calibration framework built on top of a frozen base VLA. BAS-VLA adopts a breaking-centered calibration core as the default path, and introduces a selective evidence-gated preserving auxiliary that activates only when nuisance variation is detected while task semantics remain consistent. On the OpenPI-pi0.5 / LIBERO-Object Milk-Swap benchmark, BAS-VLA maintains high success on clean (98.0%) and semantics-preserving conditions (97.5%), while reducing clean-criterion success to 0.0% under deliberate target-object swaps, demonstrating strong stale-task suppression and task-semantic separation. On validated style-preserving shifts, it improves success from 42% to 70% without degrading clean performance. These results highlight that reliable VLA behavior requires moving beyond appearance robustness toward explicit task-semantic action calibration.

Attosecond charge migration (CM) is an ultrafast process occurring when a molecule is irradiated by ultrashort laser pulses, creating a localized hole. The hole propagates rapidly through the molecule, generating electric currents and transferring charge across the molecular backbone. CM is a key ingredient in solar energy conversion, photosynthesis, and radiation damage. Despite its importance and intensive research, the fundamental physical and chemical mechanisms of CM remain not fully understood. Especially, a deeper insight into the role correlations play in the dynamics, and which chemical attributes determine CM timescales, is needed. Here we study with \textit{ab-initio} time-dependent density functional theory CM in the benchmark molecule, BrC$_4$H. We thoroughly explore CM under different initial conditions at the electronic and structural levels, including with theories of varying degrees of electronic correlations. We uncover a universal behavior where the hole moment (connecting to experimental observables) dominant frequency is roughly independent of all of these characteristics. In contrast, the timescales of the hole density evolution do vary with the chemical conditions and level of correlations. Employing a semi-analytical theory that reconstructs the hole moments in the cationic reference frame, we show that the attosecond timescale of CM is determined by molecular dipoles that filter out specific frequency responses with an analogy to optical selection rules. Our results provide essential insight into CM physics, which should be useful for interpreting attosecond experiments and engineering CM timescales by tailoring transition dipoles.

Contact-rich manipulation, requiring robots to regulate not only motion but also how they yield to external forces, has emerged as the next frontier for Vision-Language-Action (VLA) models. However, existing VLAs output purely kinematic commands, degrading performance on real-world contact-rich tasks. In this paper, we introduce CompVLA, a unified VLA framework that jointly predicts motion and stiffness matrix from RGB and language inputs. Our approach augments the conventional architecture with a dedicated Compliance Expert, which outputs time-varying stiffness and virtual displacement profiles executed via geometric impedance control. We demonstrate that CompVLA achieves the highest average success rate across diverse contact-rich tasks, outperforming both vanilla and compliance-aware VLA baselines, with ablations confirming each component is essential.

Simulation-trained humanoid proprioceptive odometry faces two transfer challenges: training trajectories generated by specific robot control policies intended for deployment cover only a limited range of motions, while sim-to-real mismatch can make unconstrained predictions unreliable. We address both with Prior-Informed Odometry from Human-Motion Tracking (PRIMO). On the data side, we generate odometry supervision by having the humanoid track diverse retargeted human motions in simulation, decoupling supervision from the deployment policies and broadening the training motion distribution. On the model side, a Prior-Informed estimator uses physics- and symmetry-informed priors to structure velocity and rotation prediction and a coarse raw-context pathway to preserve sensor context alongside encoded features, thereby strengthening sim-to-real generalization. Under a unified real-robot protocol, PRIMO reduces mean error by 31.6%-61.7% relative to the strongest evaluated external baseline in each domain-metric comparison. Across two locomotion-policy revisions, policy specialists exhibit symmetric crossover, whereas Tracking-Locomotion training reduces mean opposite-policy simulation error by 86.8%-94.6%. On real dynamic motion, Tracking-Locomotion training reduces mean error by 69.2%-81.7% relative to training on the union of both deployment policies. Across the tested motion compositions, the Prior-Informed estimator consistently lowers mean trajectory errors relative to its Unconstrained counterpart in both simulation and real-robot evaluation. Code is available at https://github.com/Agibot-Spatial-Intelligence/PRIMO.

Retrosynthetic planning advances through decisions that reshape the remaining chemical problem: a locally plausible disconnection can leave precursors whose chemoselectivity constraints complicate the rest of the route. Existing planners often channel model proposals through search or template procedures, leaving open whether a general-purpose large language model (LLM) can itself sustain and revise route strategy. We developed Rachel, a stateful environment that executes and checks LLM-directed chemistry but prescribes neither a search policy nor a stopping rule. Without supplied reference routes or route-level solutions, GPT-5.5 achieved strict closure for 111 of 120 PaRoutes120 targets and 24 of 25 targets in the separate RF25 difficult-target cohort. RF25 was drawn largely from studies published after GPT-5.5's reported knowledge cutoff. Closure required complete routes and independent source resolution of every terminal precursor after planning. On a shared PaRoutes subset, forward-model support exceeded that of most comparator methods, and Rachel received the highest mean overall route score from both method-blinded LLM evaluators. Recorded trajectories showed continued model-proposed chemistry, with revised strategies carried into subsequent steps. Replacing LLM route decisions with fixed policies reduced strict closure to 6-15/120 despite continued local chemical execution; restricting planning support also reduced closure in RF25. Within Rachel, a general-purpose LLM coordinated successive chemical choices and revised its strategy as earlier decisions reshaped the remaining problems.

Mesh texture compression typically relies on 2D UV atlases, whose chart discontinuities and mapping overhead can limit coding efficiency. To tackle this challenge, we introduce TexF, a surface-aligned texture field that organizes texture attributes in sparse voxels derived from the mesh surface. This representation supports high-resolution textures while preserving local 3D correlations for compression and enabling direct surface queries. For bitstream compression, TexF reuses established 3D attribute codecs, with voxel locations reconstructed from the decoded mesh without separate transmission. For GPU-resident compression, we develop 3DNTC, which combines quantized hash features with a lightweight decoder for random-access reconstruction at surface positions. Differentiable rendering enables image-space refinement of both voxel attributes and compressed neural fields. Experiments on the MPEG and AOM mesh compression benchmarks demonstrate improved average rate-distortion performance over representative UV-based methods for both bitstream and GPU-resident compression. 3DNTC also supports real-time rendering.

Long-video understanding must capture transient visual evidence under strict token budgets, yet existing methods compress frames, append memory tokens, or alter internal key-value (KV) caches. We introduce Prefix-Steered Recurrent Memory (PREM), a memory-token-free framework for frozen vision-language models (VLMs). PREM separates video ingestion from query answering: a recurrent writer distills visual streams into a compact 256 KiB multi-slot associative state, while a question-conditioned readout adds memory-derived key/value (K/V) steering modulations to existing non-visual prompt prefixes during prefill. This enables write-once, query-many inference without extra prompt tokens or decoding recurrence. Across six long-video benchmarks in offline and streaming end-of-stream settings, PREM consistently outperforms frozen baselines at every evaluated visual budget. Under a constrained budget of 16 frames, PREM improves macro-average accuracy by 3.06% on Qwen2.5-VL-3B, with gains of 11.0% on action antonym identification and 9.9% on localized needle retrieval. These gains require tuning 0.24% of backbone parameters at 0.03 GiB of peak GPU memory overhead.

Prompt learning efficiently adapts vision-language models (VLMs) to downstream tasks, but gains on seen classes often come at the expense of generalization to unseen classes. To address this limitation, we propose prompt ensembling with training-free routing (PETR), whose key innovation is a carefully designed dual-prompt architecture: two complementary prompts are learned from different data and objectives to emphasize seen class discrimination and unseen-class generalization, respectively. During training, both prompts are fine-tuned using a shared frozen CLIP backbone, and statistical information is collected from the training set logits. At inference time, we determine the similarity of each test sample to seen data, and route the sample to the most appropriate prompt branch. To the best of our knowledge, this is the first prompt tuning framework that performs training-free adaptive routing based on statistical similarity. This design provides an interpretable routing signal and avoids common MoE-style routing pathologies, such as router training instability and load imbalance. Extensive experiments on 11 benchmark datasets demonstrate that our framework consistently outperforms previous methods on both seen and unseen classes, achieving new state-of-the-art results.

Phase change materials (PCMs) can enable non-volatile optical memory through the large refractive index and extinction coefficient contrast between crystalline and amorphous states. Among them, $\mathrm{Sb}_{2}\mathrm{Se}_{3}$ combines low optical attenuation at telecom wavelengths, making it promising for low-loss programmable phase shifters. Its non-volatility is particularly useful for cryogenic systems, where power dissipation and thermal load constrain scalability. However, its optical properties at cryogenic temperatures remains poorly understood. Here, we report the first cryogenic optical characterization of $\mathrm{Sb}_{2}\mathrm{Se}_{3}$ integrated on a foundry silicon photonic platform from 4K to 300K for both crystalline and amorphous states. We observe that the magnitude of the thermo-optic coefficient decreases upon cooling, whereas the optical attenuation changes only weakly for both the amorphous and crystalline phases. We report the stability of the material upon repeated thermal cycling. These measurements provide the material parameters and stability required to design low-loss, non-volatile photonic memory elements for scalable cryogenic information processing.

End-to-end document parsers increasingly offer an optional reasoning mode for complex pages. On a 180-page entropy-stratified discovery sample with one frozen 4B checkpoint, complexity is the wrong decision variable. Reasoning lowers mean quality by 2.21 Overall at 1.54x tokens; a preregistered input-only model cannot predict its signed benefit (held-out AUROC 0.47, indistinguishable from chance). The benefit concentrates on pages whose ordinary pass has already collapsed, and they do not look complex: shared collapses have lower layout entropy than healthy ones yet consume 19x the tokens as degenerate repetition that doubling the budget does not cure. Switching modes rarely repairs them: 83% recur under reasoning. We instead detect collapse from the ordinary-pass trace, decompose the page by projection, and re-parse each region. Repair gains 1.40 Overall (95% CI [0.68, 2.16]) at 1.13x tokens, replicates across three checkpoints, and, with all parameters frozen, gains 2.41 (CI [1.64, 3.46]) on the remaining 1,175 benchmark pages.

Large audio-language models (LALMs) are increasingly used for a broader range of audio reasoning tasks. These models typically incorporate audio representations into a large language model (LLM) backbone to enable multimodal reasoning. Recent test-time reinforcement learning (TTRL) methods further improve LLM reasoning capability by leveraging unlabelled test data after pre-training. However, the importance of the perceptual capability of LALMs remains underexplored, particularly how much acoustic evidence is integrated and relied upon during reasoning, and how this contributes to final task performance. This gap limits the development of effective post-training methods like TTRL for audio reasoning. In this work, we first analyse how audio information is integrated and utilised during reasoning process. We quantify layer-wise perceptual reliance and show that stronger acoustic reliance is associated with higher accuracy and a larger performance gain attributable to the audio input. Building on this, we propose Perception-Grounded TTRL (PG-TTRL), which aligns label-free test-time optimisation with perceptually grounded reasoning, encouraging the model to structure its reasoning more strongly on the audio input. Experiments across LALMs and benchmarks show that PG-TTRL consistently improves reasoning performance over both the base models and standard TTRL, showing the value of perceptual-grounding optimisation for test-time audio reasoning.

Sink-aware deployment may identify important first-token attention heads before a model is quantized, then reuse that map at the edge. We test when this shortcut is safe for 4-bit NF4 weight-only post-training quantization (PTQ). Our Sink Topology Consistency (STC) metrics separate global rank preservation, top-$k$ set overlap, and layerwise sink-mass shift, and distinguish per-input sensitivity from calibration-map transfer. Across Qwen2.5-0.5B, Qwen2.5-1.5B, and Llama-3.2-1B, global bf16-to-4-bit ranks remain high at 4,096 tokens ($ρ_s \geq 0.980$), yet top-$k$ Jaccard overlap is only 0.619-0.793, corresponding to 76.5-88.5% membership retention. The global statistic also masks local failures: terminal Qwen layers shift by 6.2-7.9x their model means, whereas Llama-3.2-1B shows low, nearly uniform drift. Under a C4-to-LongBench shift, cross-domain overlap degrades more than the within-domain precision comparison for both Qwen models, but not for Llama-3.2-1B. Matched-domain 4-bit recalibration reaches 90% of a split-half stability plateau at the smallest tested $n=8$ for both Qwen models and $n=32$ for Llama-3.2-1B, though not as a sharp threshold; for the two Qwen models, updating only selected layers does not reach the full-map stability criterion. On Jetson Orin NX, the 16-sample workload takes seconds for the two models with valid on-device sink measurements. The practical message is precise: global rankings often transfer, but discrete head sets, layer-local policies, and cross-domain calibration should be revalidated after quantization.

Reactive vision--language--action (VLA) models struggle with long-horizon manipulation when visually similar observations can correspond to different actions depending on the task stage or interaction history. We refer to this ambiguity as task-state aliasing and introduce TaskAnchor, a lightweight adapter that grounds pretrained VLAs in execution history. TaskAnchor combines history-conditioned visual refinement with a milestone-supervised task-state coordinate, a scalar representing the semantic stage of execution. These signals are injected through the native visual and language interfaces, respectively, without introducing an explicit planner or modifying the action-generation mechanism. On RMBench, TaskAnchor achieves approximately 4.9--5.5$\times$ the average success rates of the published $π_{0.5}$ and X-VLA baselines, with consistent gains on RoboMemArena and real robots. The added latency is only 2.08\,ms per action chunk for $π_{0.5}$.

Computational complexity is an inherent limitation of formal synthesis for networked systems, and decomposing the global specification into local ones relaxes this limitation at the cost of conservatism. Since the granularity of the partition governs this trade-off, it is reasonable to treat the partition as a design variable, which calls for local specifications that remain correct for every partition. To this end, this paper gives each agent a local specification, written as an assume-guarantee contract that the agent can establish from local information. We first derive a necessary and sufficient condition for these contracts to decompose the global specification under a given partition. Building on this, we then present a condition under which the decomposition is correct for every partition, so that the partition becomes a free design variable. For linear dynamics and signal temporal logic formulas with affine predicates, we further synthesize a controller for each coalition by a tube-based approach. Finally, simulations on a network of input-coupled tanks show how the choice of partition trades computational cost against conservatism.

As AI agents become increasingly capable, agent-driven robotic control is emerging as a compelling paradigm. However, prevailing vision-language-action (VLA) models and world action models (WAMs) still rely on natural-language instructions to specify manipulation tasks, an ill-suited interface for agent-driven control: referentially ambiguous, spatially imprecise, redundant with the agent's inherent language understanding, and entangling intent with execution. We present AR-WAM, a visual-conditioned, agent-ready world action model that replaces language with two complementary conditions: a visual grounding prompt (a bounding box of the target) denoting the interaction object and location, and a learnable operation token dictating the atomic skill to execute. Our compact 0.5B-parameter model, with a frozen pretrained visual encoder and no language encoder, predicts scene evolution within compact latent states while decoding actions, exposing the policy's intent through explicit, supervisable reasoning signals. A model-agnostic compatibility layer provides three primitives (detect, execute, and query) so that local VLMs or online agent APIs can drive the policy directly, with long-horizon memory and closed-loop error recovery delegated to the agent side. On RoboTwin 2.0, RMBench, and a real Astribot S1 dual-arm platform, AR-WAM matches the strongest baselines on standard manipulation (87.2% average success) and outperforms them on memory-dependent and real-robot long-horizon tasks, improving success rates by 5.9% and 36.7%, respectively, while maintaining the lowest inference latency (14.1 ms).

Coding agents operate on real repository coding tasks, and persistent memory systems promise to reuse experience across tasks. Yet existing evaluations do not show whether those systems improve executable repository work. Repository benchmarks test code changes but do not isolate memory, while memory benchmarks score recall without measuring downstream coding outcomes. We introduce VibeMemBench, a benchmark for evaluating memory systems on 111 coding targets from 90 SWE-rebench V2 repositories and 3,634 history trajectories from the target repositories. The targets follow the SWE benchmark style and cover bug fixes, feature requests, interface changes, and configuration work. An agent edits each target codebase under a declared memory condition. Executable tests decide task resolution. Each target is retained only when injected history experience improves its executable outcome in a reference setting, so every target carries a prior experience whose usefulness is verified by execution in that setting. The frozen verified experience is then transferred to five held-out solvers. Direct injection raises observed task resolution on four of them by 1.1 to 4.5 percentage points while lowering agent steps on all five. Yet when four existing memory systems must construct and retrieve experience from the same history, eleven of twelve solver and system pairings fail to exceed the matched memory-off baseline. VibeMemBench exposes the gap between the useful experience that repository history holds and the experience existing memory systems deliver for repository coding tasks.

6D object pose estimation is fundamental to robotic manipulation and automation. Recent zero-shot methods have significantly improved generalization to unseen objects, but most still rely on large-scale pretrained models with substantial GPU computation and memory demands. These requirements complicate deployment on robotic platforms where perception, planning, and control share limited computational resources, while learned intermediate representations offer limited geometric interpretability for task-specific adaptation. To address these limitations, we propose G6D, a learning-free, geometry-driven RGB-D 6D pose solver. Given an RGB-D observation, an object instance mask, camera intrinsics, and a CAD model, G6D generates pose hypotheses through template-based geometric matching and refines them using silhouette and depth consistency, forming a purely geometry-driven pose estimation paradigm. This paradigm requires neither pretrained visual models nor target-specific training and preserves interpretable geometric representations throughout pose estimation. Moreover, adjustable hypothesis counts provide flexible accuracy-computation trade-offs, while a CPU-only configuration supports deployment without GPU resources. Experiments on LineMOD and five BOP19 datasets demonstrate advanced performance. Real-world pick-and-place experiments further demonstrate G6D's applicability to robotic manipulation. The complete project is publicly available at https://ai4control.github.io/G6D-Project-Page .

Vision-Language-Action (VLA) models integrate vision-language understanding with executable robot actions, enabling end-to-end learning for robot control. However, our empirical analysis reveals that existing models exhibit severe trajectory overfitting when finetuned on limited datasets. To guide the model in effectively utilizing wrist camera information, we propose MaskVLA, a masking-based fine-tuning strategy. By randomly masking a small portion of the main camera's visual information, the model is guided to autonomously learn more fine-grained, task-relevant, and effective visual features. This process leads to the emergence of robust policies, thereby enhancing the model's capability to tackle complex manipulation tasks and improving its generalization performance. Our method has been comprehensively evaluated on RoboTwin 2.0, achieving an average success rate improvement of 23.2% and 16.8% compared to $π_0$ and OpenVLA-OFT, respectively. Furthermore, experiments on real-world ALOHA robots also demonstrate the effectiveness of our approach.

Deploying efficient neural networks is essential in resource-constrained environments, yet compact models often sacrifice interpretability - a critical in safety-critical domains such as autonomous driving and medicine. This study investigates whether Knowledge Distillation transfers the spatial feature attribution of a large teacher network to a compact student. To assess the influence of the KD scheme on interpretability, we distill a ResNet-152 teacher into a ResNet-34 student on ImageNet-1K across five configurations by systematically varying the distillation temperature and soft-label loss weight. Models are evaluated on top-1 accuracy, along with two interpretability metrics: Relevance Mass Accuracy and Relevance Rank Accuracy. These metrics are computed via Grad-CAM heatmaps benchmarked against ground-truth object masks. Our results show that top-1 accuracy ranges from 71.6% to 74.0%. For Grad-CAM, RMA ranges from 7.7% to 9.7% and RRA from 7.3% to 10.1%; for Guided Grad-CAM, RMA ranges from 16.1% to 18.6% and RRA from 15.9% to 21.5%. Interpretability proves far more sensitive to the soft-label weight than to the temperature: keeping the student anchored to hard labels preserves both accuracy and coarse localization, whereas weighting the teacher heavily degrades both. Fine-grained attribution, however, fell below the undistilled baseline in every configuration tested, indicating that logit distillation transmits where a model attends more readily than the pixel-level structure of that attention. We evaluate 12 cross-architecture combinations of convolutional and transformer-based models, revealing that the inheritance of fine-grained spatial reasoning is fundamentally bottlenecked by the student's intrinsic structural biases. To our knowledge, this is the first application of this interpretability-aware evaluation framework - previously used for neural network pruning - to KD.

Low-cost planar air-bearing testbeds have matured into a standard proxy for free-flying spacecraft GNC, but they remain largely thruster-only and are rarely equipped for contact-rich, inertia-coupled manipulation. Building on the open-source ATMOS testbed, we contribute a reaction wheel and two force/torque-sensed robotic arms (LEVION) with interchangeable end-effectors, integrated as first-class control actuators through a unified ROS 2 abstraction layer. On top of the software stack we build a reinforcement-learning training environment and digital twin, and a controller that exploits these added degrees of freedom, letting classical optimal controllers and learned policies be swapped on the same hardware without modification. We validate the integrated system, PINGU, across four benchmark tasks: point-to-pose navigation (classical LQR vs. sim-to-real PPO), dynamic disturbance rejection under arm-induced center-of-mass shifts, reaction-wheel momentum stabilization, and force-controlled docking. The results show that these additions extend an ATMOS-class emulator into the contact-rich regime and bridge classical optimal control and reinforcement learning on one reproducible platform.

Agents may need capabilities exposed through external agent endpoints or service APIs. When a requester is not already bound to a provider, it must discover advertised capabilities matching its task and interface requirements. Over exact-key distributed hash tables (DHTs), broad retrieval transfers large candidate lists, whereas selective retrieval may miss relevant providers or require more replication and lookups. Open publication also lets providers inflate their exposure unless publication bounds are enforceable. We present SemDHT, a certified semantic index for discovering agent-accessible capabilities over exact-key DHTs. A two-layer semantic sketch uses coarse cells to group nearby descriptors and residual codes to narrow candidate selection. Providers publish at a bounded set of derived keys, while requesters probe precision keys before broader recall keys within a lookup budget. Anchor committees certify each descriptor's publication-key set, enabling storage services and requesters to enforce descriptor-to-key consistency. On real API descriptors and task queries, SemDHT achieves recall@10 of 0.955 against exact embedding-space neighbors and 0.947 against ToolBench relevance labels. On a corpus with controlled density augmentation, it matches the candidate exposure of tuned locality-sensitive hashing (LSH) over a DHT at recall 0.95 with 7.7x fewer lookups and reduces publication fan-out from 16 to 10. A Go/libp2p prototype deployed on same-region and cross-region 200-peer cloud overlays replays 299 Internet queries. With parallel probes and cold certificate caches, SemDHT achieves mean completion-time speedups of 3.64x and 4.11x over LSH, respectively.

Language-based 3D localization retrieves the point-cloud submap containing a target position from descriptions of nearby objects and their spatial relations. Existing methods typically compress queries and submaps into global descriptors, potentially obscuring object-level semantics and cross-description spatial coherence. We propose Position-Conditioned Evidence Localization (PosEviLoc), a query-position-aware framework for coarse text-to-point-cloud localization. Instead of relying on global matching, PosEviLoc evaluates each candidate submap using explicit semantic and spatial evidence. It models direction as a relation jointly determined by an object position and a hypothetical query position. The resulting Query-Position Spatial Evidence Field (QSEF) measures the fraction of query descriptions supported at each hypothetical position, explicitly capturing their agreement without using the ground-truth query pose to construct the evidence field. A Multi-Level Evidence Readout (MER) summarizes this evidence in a compact representation, which a lightweight MLP converts into a retrieval score. Across five benchmarks, PosEviLoc outperforms MNCL by an average of 17 percentage points in Recall@1. When used as a plug-and-play reranker, it improves MNCL by an average of 16 percentage points. Moreover, PosEviLoc introduces substantially fewer parameters and achieves faster inference speed than existing methods.

Multimodal classifiers can converge to modality-dominant solutions in which one modality dominates the joint prediction, suppressing the learning of others. Existing balancing methods mainly adjust losses, gradients, or modality contributions, largely treating modality imbalance as an optimization problem while implicitly treating the weak modality as under-optimized but representationally intact. In this work, we find that this assumption does not always hold, as persistent modality dominance can induce a representation-level collapse of the weak modality, which we term \emph{manifold modality collapse} (MMC). MMC manifests as a coupled geometric degradation in which weak-modality representations collapse onto fewer directions within each class and become less separable across classes. Motivated by this observation, we propose \emph{GeoBalance}, a geometry-aware framework that monitors these two geometric properties and reconstructs the weak modality representation only when it exhibits signs of MMC. Once triggered, GeoBalance uses a fixed Simplex-ETF class scaffold and spectral regularization to restore class separation while preventing collapse onto a few feature directions. To preserve reconstruction during joint training, asymmetric gradient projection removes the joint-gradient component conflicting with reconstruction, leaving non-conflicting optimization unchanged. Extensive experiments across six multimodal benchmarks demonstrate great improvements over competitive balancing methods, validating its effectiveness.

Neural operators have emerged as powerful surrogates for solving partial differential equations (PDEs), yet their reliability under distribution shift remains a critical barrier to deployment. Existing approaches to out-of-distribution (OOD) generalization in operator learning are largely empirical and black-box: they report aggregate error metrics without explaining why errors arise or when they will grow. We propose a structure-preserving framework that makes OOD generalization predictable and auditable. Our key idea is to parameterize the learned solution operator as a spectral filter $h_θ(λ)$ acting on the eigenvalues of the underlying elliptic operator, implemented via Chebyshev polynomial expansions and trained with a weak-form objective. This parameterization admits an exact decomposition of the energy-norm error into two observable components: a model-dependent spectral approximation term and a distribution-dependent spectral weighting term induced by the input. From this decomposition we derive three diagnostics: a conservative in-band supremum $\vareps_{\mathrm{sup}}$, a global RMS proxy $\vareps_{\mathrm{rms}}$, and a sample-dependent effective metric $\vareps_{\mathrm{eff}}(f)$. These diagnostics can be computed without access to ground-truth solutions. Through four controlled experiments, we show that $\vareps_{\mathrm{eff}}(f)\|f\|$ consistently predicts energy error under in-distribution, in-band spectral shift, out-of-band tail, and compound shifts, whereas global metrics can be systematically misleading. Our framework shifts OOD assessment of neural operators from black-box benchmarking to operator-structure diagnostics, providing a practical route to auditable scientific machine learning.

Structural causal models for time series recover contemporaneous and lagged effects, but most methods require complete observation windows and become misspecified when samples are missing. We introduce ITSY, the first continuous-optimization method for causal discovery from irregular time series under a linear model. ITSY reformulates the structural equation so that prediction uses the nearest available history rather than the possibly missing current slice, and jointly imputes missing values while learning both graphs. A weighted reconstruction objective corrects the noise transformation induced by this reformulation. Across synthetic regimes varying missingness, scale, graph density, and noise, and on a real world benchmark, ITSY consistently improves graph recovery over representative SCM-based baselines, demonstrating the effectiveness of the proposed method. The results establish a focused solution for irregular linear first-order dynamics and clarify the assumptions required for nonlinear or higher-order extensions.

Large language model agents are typically deployed with predefined configurations, although the required model capability, context, tools, permissions, memory, and computational resources can vary substantially across tasks. This study develops and evaluates AgentBetta, an adaptive AI Nano-Agent framework that represents these factors as an executable configuration and updates them through verification-driven diagnosis, selective expansion, and verification-based counterfactual contraction. The evaluation distinguishes controlled mechanism validation from external agent comparisons. On the AB-ConfigBench benchmark, AgentBetta achieved 91.38% verified success while reducing median context allocation from 64,000 to 8,000 context characters and median tool exposure from five tools to zero compared with the fully provisioned configuration. The configuration-deficiency diagnosis achieved a macro-F1 score of 0.819 with precision of 1.000 across the evaluated dimensions, and selective expansion avoided unnecessary changes to unrelated configuration dimensions. Post-success contraction preserved verification outcomes in 56.41% of evaluated one-dimension contraction probes, indicating that some successful configurations contained removable capability under the tested conditions. External evaluations indicate that adaptive configuration can improve the balance between verified task completion and capability exposure; however, the results vary across benchmarks and agent families. In particular, the cross-family replication did not reproduce the primary-backbone accuracy ordering, and specialized systems remained advantageous for certain task domains. These results support interpreting AgentBetta as a configuration-adaptation mechanism that regulates capability allocation and inference expenditure rather than as a universal replacement for specialized agent architectures.

Novel view synthesis is a key task for dynamic scene reconstruction, where high rendering speed is essential for applications such as virtual reality. Existing deformable Gaussian Splatting methods achieve high-fidelity dynamic scene modeling, but still face limitations in memory usage and rendering efficiency due to the large number of redundant Gaussians. To address these challenges, we propose Geometry-Aware Redundancy Optimization (GARO), a unified redundancy measurement framework in the adaptive density control stage of the traditional dynamic scene reconstruction pipeline. This framework first selects low-gradient candidates using an optimization activity assessment strategy, and then evaluates geometric complexity through low curvature analysis to further filter and prune redundant points, resulting in a compact and expressive Gaussian representation. Extensive experiments on synthetic and real-world datasets demonstrate that GARO achieves robust trade-offs between quality and speed, with PSNR remaining stable and rendering speed improved by 2x, validating the efficiency and effectiveness of GARO.

Task-oriented grasping (TOG) requires robots to grasp functional parts of objects (e.g., the handle of a mug for pouring), yet these affordance regions are frequently occluded in cluttered scenes. Active perception via next-best-view (NBV) planning can resolve such occlusions by moving the camera for more informative observations. However, existing NBV methods typically optimize viewpoints for grasping the target object as a whole without distinguishing which part is task-relevant. A naive adaptation, fully scanning the target object before predicting the affordance, wastes most of the viewpoint budget on task-irrelevant surfaces (e.g., the mug body for pouring). To address this, we propose ATAP, an Affordance-Targeted Active Perception framework that shifts viewpoint planning from exhaustive target scanning to targeted affordance verification. ATAP hypothesizes the occluded target geometry via a generative shape prior and predicts the affordance distribution over the imagined complete surface. In cluttered scenes, severe occlusion can make the location of the hidden affordance ambiguous, leaving multiple locations plausible given the partial observation. ATAP therefore introduces an uncertainty-aware viewpoint planner that jointly optimizes expected entropy reduction over these competing hypotheses and expected affordance verification gain from real observations. This process iterates until the affordance is sufficiently verified for grasp execution. Experiments in simulation and real-world cluttered scenes show that ATAP substantially improves the functional grasp success rate over fixed-view TOG baselines, and outperforms reconstruction-based active perception with over 57% fewer NBV steps.

Physiological time series such as electrocardiograms (ECG) and electroencephalograms (EEG) exhibit complex temporal structure, substantial acquisition variability, and a strong need for transparent decision-making. Although deep models can achieve high detection performance, they often provide limited insight into why a segment is anomalous, how local anomalies relate over time, and whether a detection belongs to a broader recurring pattern. We propose Signal2Symbol, a neuro-symbolic framework for explainable biosignal anomaly detection. The method first converts ECG/EEG signals into symbolic sequences using either a learned VQ-VAE (Vector Quantized Variational Autoencoder) codebook or a SAX (Symbolic Aggregate approXimation) baseline. It then constructs bigram enriched token-window transactions and scores anomalies through rare itemset evidence derived from minimal rare itemset mining. Detected anomalous windows are merged into intervals and related using Allen interval algebra, enabling composite temporal explanations such as escalation chains, artifact overlap, and cross-channel synchrony. Finally, we introduce a rare temporal concept lattice based on Formal Concept Analysis (FCA), which groups anomalous intervals by shared rare symbolic evidence, Allen temporal relations, channel context, and robustness attributes. The resulting Galois lattice compresses many local detections into interpretable families of temporal-symbolic anomalies. We evaluate on three public benchmarks: MIT-BIH Arrhythmia (beat-level ECG), PTB-XL (record-level ECG), and the Bonn EEG dataset (segment-level EEG). We stress-test robustness under additive noise and baseline-wander perturbations. The results highlight the value of neuro-symbolic tokenization for temporal anomaly analysis and show that Allen/FCA reasoning provides compact, interpretable summaries of local detections.

Failures of high-resolution MLLMs are commonly attributed to a visual problem, motivating zooming, cropping, and related visual interventions to recover fine-grained evidence or suppress interference. Yet recent studies suggest that relevant visual evidence is already encoded in intermediate representations, indicating that visual-side improvements alone insufficient. This raises a natural question: does the remaining bottleneck lie in the text that guides visual search? We identify a previously overlooked linguistic bottleneck: questions formulated for answering do not necessarily specify the visual evidence required for localization. To address this mismatch, we introduce EviSpec, a training-free compiler that derives complementary evidence specifications while preserving the original question for final reasoning. We further validate it through matched-control experiments that isolate the roles of evidence specification and localization. With the search budget fixed, structured evidence specifications yield an 8.6% relative gain over generic requests. With evidence geometry matched, the evidence localized by EviSpec yields a 14.8% relative gain over random evidence. Together, these controls isolate the benefit of specifying what evidence to seek rather than merely expanding visual access. Across all five MLLMs, EviSpec consistently improves upon the corresponding baseline on each of the three benchmarks, yielding average relative gains of \textbf{10.4%, 8.8%, and 12.4%} on V\textsuperscript{*}Bench, HR-Bench-4K, and HR-Bench-8K, respectively. Beyond high-resolution reasoning, EviSpec also achieves state-of-the-art performance on VQA and hallucination-focused benchmarks.

Perception for embodied agents is video-based, often multi-view (ego, exo, or both), and inherently continual, with simultaneous task and viewpoint shifts. Yet continual learning (CL) remains dominated by exo-only recognition tasks, obscuring behavior under these real-world coupled shifts. We introduce Continual Ego, E}xo, and Ego-Exo Learning (CE$^4$L), a unified multi-view CL benchmark spanning four representative tasks: cross-view referenced skill assessment, temporal action segmentation, cross-view association, and action anticipation & planning. CE$^4$L highlights challenges largely absent in prior CL benchmarks, including cross-view correspondence, view-dependent asynchrony, and heterogeneous semantic objectives. To this end, we propose Video Incremental Subspace-routed Task Adapters (VISTA), a parameter-efficient baseline method that stores task-specific updates in lightweight adapters and performs training-free routing via residual distance to task-specific whitened subspaces estimated from second-order statistics. Extensive experiments demonstrate the significantly varied efficacy of representative CL methods across CE$^4$L settings, while VISTA is consistently competitive and achieves state-of-the-art overall performance. Our source code for benchmarks and methods is available at https://github.com/AnAppleCore/CE4L .

We present Elevator-VIGS, a visual-inertial 3D Gaussian Splatting SLAM system that keeps tracking and mapping through elevator rides. Inside a moving elevator, the two sensors are in conflict. The camera sees only the robot's motion relative to the elevator, while the IMU senses that motion plus the elevator's motion relative to the world. This conflict is challenging for existing visual-inertial estimators. If vision dominates, the estimator tracks only the robot's motion within the elevator and misses the elevator's rise, and if the conflict remains, the estimator diverges. We observe that the conflict comes from forcing both observations into a single coordinate frame. We instead estimate the robot's pose in the elevator's coordinate frame, and the elevator's motion relative to the world as a per-keyframe transport state, the elevator's rise and vertical velocity, within dense visual-inertial bundle adjustment. Elevator-VIGS detects rides zero-shot with a vision-language model and a depth network, and constrains the transport state at the departure and the arrival. We record real-world and simulated elevator sequences. On these sequences, Elevator-VIGS achieves state-of-the-art tracking and rendering performance. On four elevator-free public benchmarks it keeps the state-of-the-art performance of VIGS-SLAM. Project page: https://ruizhou-cn.github.io/elevator-vigs/.

Large language model (LLM) agents increasingly execute multi-step workflows through tool use and interaction with users and environments. However, current agent evaluations are largely English-centric, limiting our understanding of agent capabilities in multilingual settings. We introduce BabelFlow, a benchmark-general agentic workflow that adapts existing agent benchmarks to new languages by analyzing runtime dependencies, coordinating structure-preserving translation, and combining multi-layer verification with human review to preserve task and evaluation semantics. Using BabelFlow, we construct BabelArena, a task-aligned benchmark comprising 16,146 instances derived from 702 canonical tasks across four benchmark families, 13 domains, and 23 languages. Experiments with five frontier models show that no single model dominates across benchmark families and that cross-language disparities extend well beyond task success. Lower-resource languages exhibit distinct failure patterns, with larger shares of tool-use and control-flow errors rather than answer-quality errors alone, pointing to gaps in reliable task execution across the resource levels of these languages. On the same tasks, agents in low-resource languages also consume substantially more tokens than in English (up to roughly twice the input) without proportional increases in interaction length, and language consistency degrades further on tasks requiring structured output, where switches are directed overwhelmingly toward English. We believe BabelArena provides a foundation for advancing research on reliable and efficient multilingual agents.

End-to-end autonomous driving maps current observations directly to future trajectories, yet those trajectories must remain valid as the scene evolves. Future state modeling aims to address this temporal mismatch, but general representations often contain information unrelated to ego planning and affect trajectory generation only through auxiliary supervision, static conditioning, or proposal evaluation. We propose FeasibleFlow, a one-step end-to-end generative framework that jointly transports a configuration-space feasibility field and multimodal ego trajectories. Our Asymmetric Joint MeanFlow uses the pathwise Jacobian-vector product in the MeanFlow identity to incorporate field evolution into trajectory transport. Because safety feedback is sparser than progress feedback, we further introduce the Anchor-relative ranker (ARR) and Pareto-ReinFlow to balance safety and progress in candidate selection and generation, respectively. Experiments on the NAVSIM benchmark demonstrate the strong performance of FeasibleFlow and validate both the joint transport of feasibility and trajectories and the proposed safety-first mechanisms.

Robots operating in human-centered environments are typically designed to execute explicit instructions, and most robot-learning datasets likewise pair observations with task instructions or low-level actions. Although recent work has begun to explore proactive embodied assistance, existing resources target different settings and action levels, leaving real-world human-centered multimodal decision-making underexplored. We formulate this problem as \textit{Proactive Robot Action Reasoning} (\textit{ProRobo}), an upstream cognitive decision problem in which a robot must determine which action to take based on multimodal human and environmental cues without explicit action instructions. To support ProRobo, we introduce \textit{ProAction}, a real-world multimodal dataset containing 10K samples of visual observations, audio signals, and text inputs across 12 daily-life scenarios in five common scenes. To construct cognitively grounded high-level action supervision, we develop a two-stage human-in-the-loop pipeline that combines appraisal-guided candidate generation with Affective Theory-of-Mind-guided human refinement, explicitly incorporating contextual judgment about human states, urgency, feasibility, and potential risk into action annotation. Based on this supervision, we benchmark representative Multimodal Large Language Models (MLLMs) and introduce \textit{MMC2Act}, a reference model that implicitly learns the mapping from multimodal observations to cognitively grounded high-level actions. Experiments across modality settings, subject-disjoint generalization, cross-dataset transfer, and human evaluation show that general-purpose MLLMs struggle with proactively reasoning high-level actions from multimodal cues, whereas training on \textit{ProAction} substantially improves performance.

Humanoid loco-manipulation faces two prominent limitations: controllers using continuous velocity commands cannot precisely regulate individual footholds, while specialized foothold-tracking modules are difficult to integrate with whole-body manipulation. Furthermore, standard action-based imitation distillation primarily transfers expert actions, without explicitly encouraging a shared representation of heterogeneous skills. This paper introduces STRIDER, a hierarchical multi-gait framework to bridge these gaps. The framework integrates terrain-aware 3D stepping logic, Adversarial Motion Priors (AMP)-based natural walking, and Cartesian upper-body control: its stepping expert selects feasible footholds in the stance-foot frame and generates clearance-aware swing trajectories. To fuse distinct walking and stepping experts into one executable student policy, we propose Latent Distillation Proximal Policy Optimization (LD-PPO), a distillation algorithm augmented with teacher-conditioned latent alignment. By jointly optimizing on-policy reinforcement learning, DAgger-based action reconstruction, and latent alignment, LD-PPO transfers expert actions while encouraging a shared skill representation across heterogeneous modes. Simulation and real-robot evaluations on the TianGong Omni humanoid show that LD-PPO outperforms vanilla distillation-PPO in foothold-tracking and posture-tracking accuracy. Deployed on hardware, STRIDER realizes multi-gait loco-manipulation with accurate foothold and end-effector tracking.

Latent action models infer a code for the transition between two frames of action-free video. Recent methods regularise this code to compose additively and reverse antisymmetrically, and report order-of-magnitude reductions in the resulting errors as a label-free certificate that the code has captured temporal structure. We show that this conclusion does not follow. Reconstruction drives the decoded transition toward a difference of state features, for which both identities hold for any pairing, a solution the metric cannot distinguish from one encoding nuisance state or a coordinate convention. Across five source domains, a trained but unconstrained counterpart already achieves 83-97% of the reduction relative to an untrained anchor. The residual fold is governed as much by the decoder family as by what is learned. A constrained model retrained after its temporal pairing is destroyed still reaches, in each domain, a lower error than the unconstrained model on real data. Downstream, preserving the temporal pairing yields no consistent advantage on LIBERO-GOAL or LIBERO-SPATIAL, and across the tested arms the code's mean linear action decodability falls as the algebraic error improves. We also test the most direct repair, a violation-contrastive objective that requires the algebra to fail on destroyed pairings: in the tested configurations it yields only a marginal separation within the reconstruction budget, on training and test triples alike. We recommend a validation protocol that these methods currently lack: a baseline-corrected metric, retraining on destroyed pairings, and a seed-budget analysis.

Long-running LLM agents require memory that persists and evolves across sessions. Text-based memory retrieves and reconstructs past interactions at every query, making long-horizon performance increasingly dependent on retrieval quality and contextual reasoning as histories grow. Parametric memory encodes experience directly into model computation, but existing approaches provide limited support for cross-session memory evolution. Their coupling to a specific backbone further restricts memory reuse after model replacement. We introduce RPMem, a two-stage architecture that compiles each session into a model-independent latent memory through forward computation and selectively integrates it with retained memory via a task-trained recurrent gate. The consolidated memory is then mapped to backbone-specific low-rank adaptation (LoRA) parameters, allowing the encoding capability to transfer when the backbone is replaced. Evaluation across three long-term memory benchmarks and five diverse backbones demonstrates broad generalization with near-constant update cost and memory footprint. With Qwen3-8B on PERMA, RPMem reaches 85.52%, outperforming the strongest parametric and text-based baselines by 5.32 and 12.98 percentage points, respectively. Ablations validate the complementary roles of session compilation and cross-session consolidation, while dynamics analyses reveal that the gate acquires task-specific memory integration strategies. These results establish RPMem as a lifecycle-independent parametric memory framework that maintains evolving cross-session memory that remains reusable across backbone replacements. Our implementation is available at https://github.com/Quark-Medical/rpmem/tree/main.

Long-horizon conversational memory is especially challenging in multi-actor settings, where relevant evidence is distributed across participants and contexts and previously established information may later be revised. We introduce EGMEMORY, which formulates long-horizon multi-actor memory as a searchable state machine that separates persistent message-level evidence from an explicit active state. At write time, adaptive state resolution and an evidence-grounded propose-verify-commit protocol govern how this state evolves. At read time, adaptive evidence navigation iteratively resolves the state and supporting evidence required for a query, using conversational structure to narrow the search space and lexical-semantic relevance to rank candidates. The system operates through prompting and tool use without memory-specific policy training. EGMEMORY achieves 68.2% on GroupMemBench and 77.9% on EverMemBench, outperforming the strongest evaluated baselines by 22.7 and 21.4 percentage points, respectively. It further reaches 73.6% on the dyadic LoCoMo benchmark, demonstrating generalization beyond multi-actor conversations. We will release the codebase upon formal publication.

Reinforcement Learning with Verifiable Rewards (RLVR) is expanding from tasks with well-defined correctness signals, such as mathematics and code, toward multifaceted quality requirements specified by multi-dimensional rubrics. Since policy optimization consumes one scalar per rollout, rubric-based pipelines must map multiple criterion scores into a scalar reward. This aggregation is often treated as score scaling, but it implicitly determines how quality dimensions trade off during training. The prevailing practice, normalizing each criterion and taking a linear combination, assumes that cardinal score differences are comparable across criteria and that gains on one criterion compensate for failures on another; both assumptions are unreliable when criteria are semantically heterogeneous. We propose Reinforcement Learning with Verifiable Rubric-based Ranking (RLVR$^2$), a verifiable ranking paradigm for rubric-based RLVR. For each criterion, RLVR$^2$ converts rubric scores into criterion-specific within-group ordinal outcomes, recovers a latent utility from the resulting comparison matrix, and merges these utilities into one training signal. By retaining only within-group ordering and discarding raw score magnitudes, RLVR$^2$ avoids calibrating heterogeneous rubric scales. It further supports objective-preserving attribute adjustment: auxiliary attributes that correlate with observed rankings but are not training objectives can enter the estimation without expanding the rubric or rewarding them directly. Across three model scales and 16 benchmarks, RLVR$^2$ consistently outperforms representative rubric-based baselines, achieving the best overall performance on most benchmarks at every scale. Analysis shows it controls systematic effects tied to reasoning efficiency and response formatting while preserving the quality objective.

Memory systems are becoming a core component of LLM agents, but constructing and maintaining memory remains expensive because it relies on repeated calls to large proprietary language models. This cost creates a major barrier to deploying memory-enhanced agents at scale. In this paper, we present Pseudo Self-Distillation (PSD), a framework that enables small language models (SLMs) to construct hierarchical memory representations by distilling behavior from a strong black-box oracle through a multi-stage training pipeline. Standard distillation methods require access to teacher logits or hidden states, which closed models do not expose. Unlike conventional self-distillation settings, where supervision is derived from a model's own predictions, sampled rollouts, or aggregated outputs, PSD enables a single-model distillation setup while channeling external oracle knowledge through the prompt. PSD uses a single small model in two roles: a teacher that sees a privileged prompt containing the oracle's answer as reference context, and a student that sees only the task prompt. The student learns to reproduce the teacher's output distribution, absorbing oracle-guided behavior into its own weights without accessing the oracle's internals. On LoCoMo, PSD-trained Qwen3-0.6B, 1.7B, and 4B match or exceed GPT-4.1-mini on downstream retrieval at a fraction of the deployment cost, with off-policy PSD achieving the strongest results across most conditions. We further show that this memory-construction capability transfers out-of-distribution to LongMemEval, despite the students being trained exclusively on LoCoMo with no exposure to LongMemEval data.

Bimanual manipulation requires policies that coordinate two arms while adapting their functional roles to scene geometry, object configuration, and task context. Learning such scene-conditioned role adaptation remains challenging, as demonstrations may contain uneven role distributions that limit generalization to underrepresented arm--role configurations. In addition, many bimanual policies predict actions in fixed left- and right-arm action spaces. While this provides a natural parameterization for robot control, it does not explicitly specify how behaviors should transform when functional roles are exchanged across arms. Across different scene initializations, the two arms may follow a similar coordination pattern, but the role-specific behavior assigned to each arm should change with the scene. Therefore, we propose BiRoAD, a Bimanual Role-Adaptive Decomposition framework for learning shared and role-adaptive representations in bimanual policies. Given bimanual trajectory or action-token features, BiRoAD decomposes these features into swap--symmetric and swap--antisymmetric components: the former captures coordination structure invariant to arm exchange, and the latter captures role-specific distinctions that vary consistently with functional role assignment. The two components are then recomposed as residual updates to the original paired arm representations, allowing BiRoAD to serve as a modular feature transformation without changing the policy inputs, imitation-learning objective, or requiring manually defined role labels. Across multiple bimanual manipulation tasks with balanced and imbalanced role distributions, BiRoAD improves robustness across role configurations over corresponding base policies, with notable gains on underrepresented role configurations.

Engineering change order (ECO) is an important step in repairing timing and electrical violations during the late stages of chip design. Existing Agentic EDA methods primarily focus on tool invocation, with less attention to model decision quality and targeted training. A central challenge in ECO is multi-round decision-making: the model must use the results of each round to determine the next repair action. We propose WaveletECO, which integrates a closed-loop execution platform with large language models to enable agents to execute ECO decisions effectively. We also train a local 9B model through supervised fine-tuning and CPO-SimPO using execution demonstrations and decision-preference data, enabling ECO decision-making with a locally deployed model. Across 594 evaluation runs on 22 designs, WaveletECO-Policy (BF16) and (INT8) score 79.63 and 79.65, respectively, compared with GPT-6 Astra's 77.44. The estimated inference cost of INT8 is about 1/147 of GPT-6 Astra's. These results show that specialized model training supports effective, low-cost multi-round ECO repair, with repair quality retained under INT8 quantization.

Diffusion models generate high-quality images, yet often violate the physical laws governing mirror reflections. Reflections often suffer from geometric aberrations, including positional offsets, directional misalignment, proportional imbalance, and structural distortion. These failures remain evident even in contemporary state-of-the-art generative systems. Existing methods itigate this problem through synthetic data scaling or auxiliary depth conditioning, yet their merely reliance on latent-space noise reconstruction losses as implicit supervision prevents direct enforcement of reflection-specific geometric and perceptual constraints. To bridge this gap, we present PhysReflect, a geometry and perception guided diffusion framework that decodes the predicted clean latent into pixel space at each training step and applies annealed supervision through two complementary differentiable objectives. The Geometric Loss enforces mirror-induced spatial consistency through sparse epipolar correspondence and dense boundary projection alignment, where a SAM2-based TwinTrack mechanism provides stable in-mirror localization for boundary-aware supervision. The Perceptual Loss preserves reflected appearance by combining Semantic Consistency Loss, which maintains reflected identity and appearance via DINOv2 features, and Lighting Consistency Loss, which regularizes depth, surface-normal, and illumination coherence under monocular geometry priors. Experiments on synthetic and real-world benchmarks show that PhysReflect outperforms prior mirror-reflection methods in geometric, perceptual, and physical-plausibility metrics, as well as qualitative visual results.

Non-prehensile manipulation is practical for relocating large, heavy, or geometrically ungraspable objects. Yet, long-horizon pushing of arbitrarily-shaped 3D objects couples three problems: 1) where to push the object so as to approach the target pose, 2) whether each push is stable and reachable, 3) whether subsequent actions remain feasible. We present an object-centric pushing policy within a feedback-guided hierarchical framework. At the low level, a learning-based policy predicts contact actions from a pose- and scale-normalized point cloud, conditioned on a near single-step subgoal. A stability score is applied to evaluate the predicted contacts by a quasi-static sliding-versus-tipping analysis. At the high level, BIT$^*$ first searches for an object path, and the next several subgoals are checked by contact prediction and robot motion planning for future feasibility. Failed motion plans, as feedback, change the local path costs and trigger re-planning. During execution, only the first feasible action is executed. In simulation, we evaluate 22 objects in six different scenes, upon which we also conduct comprehensive ablation studies. Results demonstrate that our method outperforms baselines with a clear margin and can reliably achieve long-horizon object pushing tasks under different situations. We also report quantitative real-robot experiments with a Franka arm and qualitative demonstrations with a mobile manipulator for large and heavy objects, with directly zero-shot sim-to-real transfer.

Multi-omics sequences contain complex biological patterns, yet deciphering their mechanisms for automated scientific discovery remains challenging. As large language models (LLMs) interpret these sequences, evaluating both predictions and scientific reasoning is critical. However, existing benchmarks for multi-omics sequence tasks rely on classification and regression metrics, neglecting whether models grasp the underlying biological evidence. We introduce OmicsBench, the first reasoning benchmark for multi-omics sequences, comprising 1,160 expert-validated questions across six tasks spanning DNA regulation, RNA processing, and protein function. OmicsBench requires traceable evidence chains, evaluated using instance-specific rubrics developed with domain experts. Evaluating 17 LLMs reveals an inverse relationship: while scientific LLMs outperform general-purpose LLMs in sequence classification accuracy, they fail to provide valid evidence to support their predictions. One plausible interpretation is shortcut learning: specialized models may rely on statistical patterns rather than the biological mechanisms needed for scientific discovery. Motivated by this finding, we introduce tool-augmented on-policy distillation (TA-OPD), a post-training method to align sequence prediction with evidence-grounded biological reasoning. Across five Qwen3.5 models spanning 0.8B to 27B parameters, TA-OPD consistently strengthens biological evidence grounding while improving predictive performance on most tasks. These gains persist across model scales, indicating that stronger sequence reasoning does not arise solely from increased model capacity, but can be improved through evidence-aware training. Together, OmicsBench and TA-OPD provide a framework for diagnosing reasoning failures in multi-omics LLMs and a path toward models whose predictions are better grounded in biologically meaningful evidence.

Dynamic rope manipulation is highly sensitive to unknown object dynamics: the same robot motion can produce substantially different responses across ropes, while explicitly identifying the relevant physical properties is difficult. We present RopeFormer, a history-conditioned framework that uses prior task interaction as context for subsequent control. The policy retains cross-trial action-response history while keeping its weights fixed and requires no explicit online rope-parameter estimation. In matched simulation evaluations across sustained single-arm rotation, bimanual rotation, and transient whipping, retaining context improves subsequent control relative to resetting the same checkpoint, with the benefit varying across rope dynamics and observation settings. We further deploy the frozen policies on a Unitree H1-2 with previously unseen physical ropes. From T1 to T3, target-acquisition time decreases by 30.9% for Rope Swing and 33.9% for Rope Twirl, while mean Rope Whip target hits increase from 0.2 to 2.3 out of three. These results show that prior interaction can provide effective control context for dynamic deformable-object manipulation. Robot videos, code, and data are available at https://ropeformer.github.io/.

Vision foundation models targeting Earth observation (EO) tasks are commonly evaluated on clean downstream benchmarks, but operational EO products can already contain spatial, radiometric, alignment, noise, and harmonization defects before reaching the model. Existing robustness evaluations often use generic image corruptions or broad domain shifts, which do not isolate these product-level failure modes. We introduce \textbf{RSPDBench}, a physically grounded \textbf{r}emote-\textbf{s}ensing-\textbf{p}roduct \textbf{d}egradation \textbf{b}enchmark for vision foundation models. RSPDBench evaluates five EO datasets, seven foundation-model entries, and two supervised baselines under audited primitive degradations and compound product chains. Each model is evaluated under its clean-selected native protocol, with robustness measured as the drop from its own clean baseline. Our analysis reveals that degradation sensitivity is strongly structured: resolution-conditioned and channel-grouped encoders protect different failure axes, and the same physical defect can hurt one model while helping another. Compound chains expose failures that isolated degradations do not predict, with model-dependent amplification, saturation, or component dominance, and excess drops up to $38$ percentage points beyond the strongest component. These results show that EO robustness cannot be characterized by clean accuracy or generic perturbation tests alone; it must also be measured against the structured defects that remote-sensing products carry into deployment.

Computational Fluid Dynamics (CFD) is widely used to evaluate ventilation and contaminant transport in occupied buildings, but deployment at scale is limited by three bottlenecks: acquiring room geometry without costly scanning hardware or manual CAD modeling, decomposing the scene into individually manipulable objects, and reconfiguring those objects for alternative layouts without re-capturing the room. We present a semi-automated workflow that converts a single 360-degree video of a room into individually editable, simulation-ready geometry assets. A dense point cloud is reconstructed using Neural Radiance Fields (NeRF), and 2D instance masks from text-prompted SAM 3 segmentation are lifted to 3D using multi-view consensus and depth-band filtering. Points are separated into object instances with an octree, and occlusion gaps are healed with a connectivity graph. Chair templates are fitted by Iterative Closest Point (ICP) alignment, and table geometry is generated procedurally. A browser-based editor supports quality assurance and rapid construction of alternative layout configurations. A steady Reynolds-averaged OpenFOAM solution then drives transient passive-scalar transport; the setup is verified using a mesh-sensitivity study and validated against an IEA Annex 20 benchmark. We apply the workflow to two university classrooms and a tiered lecture-hall auditorium. The capture-to-geometry pass takes two to five hours per room on a consumer workstation. In a controlled obstruction sequence in one classroom, the modeled half-clearance time varies non-monotonically as furniture is added, and a cross-room comparison indicates that clearance behavior cannot be reliably extrapolated between rooms, motivating per-room geometry acquisition. By making that acquisition low-cost, the workflow makes geometry-resolved comparative ventilation studies practical for spaces such as classrooms.

Vision-language navigation (VLN) has largely been developed for indoor and terrestrial robots, where language can often be treated as a static goal and motion is approximated by discrete or near-instantaneous actions. These assumptions break down for unmanned surface vehicles (USVs): river navigation requires continuous motion under inertia and limited maneuverability, while long-horizon instructions must be executed through sparse and visually ambiguous maritime landmarks. We introduce RiverVLN, to our knowledge the first benchmark designed for long-horizon USV VLN under continuous riverine motion, and PGT-NAV, a phase-grounded temporal navigation framework for USVs. Rather than directly mapping an entire instruction to motion, PGT-NAV converts it into an ordered sequence of visually verifiable semantic phases and maintains the active phase online through grounded visual and motion evidence. This explicit semantic progress state is fused with visual-motion history and phase-specific grounding to predict six local SE(2) pose increments. The resulting trajectory is executed in a predict-execute-re-observe loop, where the vessel executes toward W3, updates phase and grounding, and replans through a map-based safety layer. Experiments show that PGT-NAV substantially reduces recursive position and heading drift relative to GNM-style and ViNT-style baselines and achieves an average success rate of 0.79 in Unity-ROS closed-loop navigation. Unseen bridge-opening trials and real-world USV experiments further demonstrate that the phase-grounded representation transfers from controlled evaluation to physical USV deployment.

Language-guided manipulation can depend on physical properties that visible appearance does not reveal. Temperature is one such property, but object datasets for robot learning rarely associate measured temperatures with object appearance and geometry. We present HEARTH, an object-centric RGB-thermal-3D dataset of 90 physical objects from 18 everyday categories, comprising 145 captured object states. Our pipeline maps apparent surface temperatures onto reconstructed meshes through camera calibration and pose transfer. The dataset includes raw temperature measurements, camera parameters, RGB-textured meshes, and thermal textures for simulation. We use these assets to construct three LIBERO-derived tasks and collect 1,200 demonstrations for fine-tuning a pretrained vision-language-action (VLA) model, $π_{0.5}$. In an ablation study, adding thermal observations to the VLA increases success on temperature-dependent object-selection tasks from 35.0% for the RGB-only baseline to 75.0%. These results demonstrate the utility of HEARTH for training robot policies to follow temperature-related instructions.

Screen-recorded web editing requests contain weak deictic expressions such as ``this'' and ``there,'' whose referents depend on speech, cursor trajectories, page state, and edit history. Such requests require intent recovery beyond the explicit specifications assumed by many existing web-editing benchmarks. We introduce Omni2Web, a bilingual benchmark of 918 instances spanning 13,907 edit steps. It defines three complementary tracks: Direct Editing evaluates webpage editing from recordings, Instruction Recovery measures explicit intent recovery, and Instruction Utility tests whether recovered instructions can drive a fixed code executor. We evaluate 17 open- and closed-source models. The best models attain 51.17 on the Edit Fidelity Score (EFS) for Direct Editing and 49.14 on the Instruction Recovery Score (IRS); under the fixed executor, the strongest recovered instructions reach 51.08 EFS, still far below the 89.69 EFS obtained with oracle instructions. Step-level analyses show that correct grounding does not guarantee successful edits, while some Omni models recover instructions that the fixed coding model executes substantially better than their direct edits. Controlled ablations further demonstrate the value of temporally aligned audiovisual evidence, while alternative judges preserve the leader and broad ordering. Together, these findings reveal substantial headroom in multimodal intent recovery and code execution and highlight the promise of pairing Omni rewriters with coding models.

Long-form audio performance is often summarized by context length and aggregate accuracy, obscuring how language, evidence, and task jointly shape difficulty. We introduce MuLA-Bench: 5,038 open-ended questions over 1,769 in-the-wild recordings totaling 1,377.9 hours, covering 16 languages and eight domains. A balanced Language x Domain semantic track supports controlled comparisons, while a complementary acoustic track preserves naturally occurring non-speech evidence. Evidence-grounded generation, shortcut checks, and language-expert review provide auditable questions without translating a shared source set or injecting target sounds. We evaluate ten audio-language models and conduct pooled diagnostics on a fixed eight-model cohort. Language rankings change across domains and tasks; acoustic-semantic performance gaps vary with the requested operation; and temporal errors can persist after the correct event is identified. Long-range retrieval is comparatively strong, while precise clock alignment and factual grounding of natural acoustic events remain fragile. MuLA-Bench thus exposes conditional failure patterns that a single long-context score does not capture.

Socially competent humanoid robots must communicate affect and intent through gesture as well as speech, yet open-ended interaction must become motion that is both expressive and executable on a specific body. This demands semantic flexibility for contextual social intent while preserving deterministic, embodiment-aware robot control. We present EmoPose, a vision-language model (VLM)-guided framework that bridges this gap through an executable semantic interface. Given language, dialogue history, and optional visual context, the VLM selects an ordered gesture plan containing a communicative class, library variant, intensity, and speech anchor. A scalable robot-owned motion library defines the available expressive vocabulary and the source of 14-DoF joint targets. Pose Studio supports automatic trajectory generation, MuJoCo preview, and automatic synchronization of new library entries with the VLM guide; deterministic robot-side modules validate plans, construct trajectories, schedule gestures, and manage queueing and interruption. This division lets the interaction repertoire grow for new social contexts without changing the control interface or delegating raw joint commands to the foundation model. On the EmoPose-Bench, structured GPT-5.5 planning reaches $98.25\pm0.52\%$ on the Easy tier and $76.50\pm0.54\%$ overall, exceeding same-model direct-label prompting. Further tests validate dialogue-context use and ordered multi-action composition. The system completes the nominal MuJoCo suite and realizes all 29 authored variants on the physical Unitree G1. A four-stop laboratory tour demonstrates expressive narration with interruption, camera-grounded dialogue, and navigation.

Cache-based test-time adaptation improves CLIP predictions by storing and retrieving examples from the target stream while keeping the model frozen. However, existing methods largely treat the feature space used for image-image retrieval as fixed, leaving open how much adaptation depends on the retrieval space itself. We study this question by fixing the memory and changing only the retrieval encoder, finding that the same memory can yield very different gains: across sixteen retrieval spaces, ImageNet-A cache gain ranges from at most +0.44 points for CLIP and MAE to +19.7 +/- 0.4 for DINOv2-L, while label-free retrieval-space selection retains 98% of oracle gain on ImageNet-V2. These results show that memory quality depends not only on which examples are stored, but also on how they are retrieved. Motivated by this finding, we propose MARC (Memory Augmented Retrieval for CLIP), a training-free system that uses frozen CLIP for prediction and DINOv2-B for retrieval with a single fusion weight. A single-view cache repairs 1074 +/- 21 baseline errors, compared with 878 +/- 4 for a 64-view ensemble, at roughly one seventh of the cost. Across four ImageNet distribution shifts, MARC reaches a 67.91% OOD average and, at matched DINOv2-B scale and eight views, achieves 64.17 +/- 0.31% versus 62.75 +/- 0.15% for a graph-based cache system while running 2.6 times faster. Overall, our results establish retrieval space as a first-order design choice for robust cache-based adaptation in remote sensing, scientific imaging, and changing visual environments.

In frame interpolation tasks, motion ambiguity in the training set causes models to generate blurry intermediate frames. Moreover, the assumption of uniform motion between frames during inference further leads to inaccuracies in the generated intermediate frames. To tackle these challenges, we propose an Accurate motion estimation algorithm with Bézier Control point, ABC-Inter, for efficient frame Interpolation. Specifically, ABC-Inter designs an Accurate Flow estimation Module (AFM) by decoupling two-frame features and mapping to corresponding coordinates to better estimate the optical flow between the two frames. Furthermore, ABC-Inter eliminates motion ambiguity in the training set by introducing Bézier control points that are computed using the input frames and the intermediate ground-truth (gt) frames. This allows the model to estimate accurate optical flow between two frames during the training process, thereby solving the blurriness problem in the generated intermediate frames during inference. Benefiting from the more accurate flow estimation between two frames, we can introduce additional frames and directly use multiple flows to calculate Bézier control points for modeling non-uniform motion without retraining the model. Simultaneously, to realize the estimation of non-linear motion using only two frames, we also introduce a new Bézier control point estimation module which achieves better motion estimation between the two frames by performing fine-tuning on the model in the second stage. Experimental results demonstrate that our ABC-Inter achieves state-of-the-art performance on multiple benchmark datasets and exhibits excellent visual perception.

Humans can effortlessly localize the direction of a sound source and integrate it with visual cues for reasoning, yet this remains challenging for embodied agents. In particular, it is still unclear how to effectively evaluate and model spatial audio understanding in embodied settings. To address this gap, we introduce \textbf{OmniEchoBench}, a unified benchmark for spatial audio-visual perception and audio-vision-language navigation. OmniEchoBench comprises six tasks over 197 real-world spatial audio-visual scenes, 2,972 question-answer pairs, and 900 navigation samples with first-order ambisonics (FOA) audio collected from 30 real-world environments. To enable scalable training supervision, we develop a controllable rendering pipeline for spatial audio. It preserves geometric consistency among sound sources, visual observations, and agent trajectories. Building on this, we propose \textbf{OmniEcho}, a spatially aware omni-modal model. It introduces an FOA spatial encoder alongside a pretrained semantic audio pathway. Extensive experiments show that OmniEcho achieves state-of-the-art performance on spatial audio-visual perception. For our sound-guided navigation, OmniEcho reaches a performance level close to that of traditional vision-language navigation. These results demonstrate that spatial audio can serve as a valuable signal for embodied scene reasoning and navigation, while also highlighting fine-grained spatial localization and distance estimation as important open challenges. Our code and data will be available in https://github.com/PKU-VaLuE-Lab/OmniEcho/tree/main

Bayesian filtering provides a principled framework for online state estimation under uncertainty, yet its application to systems with high-dimensional states and complicated posterior distributions remains challenging. Recent generative models, such as flow matching, have shown potential in Bayesian filtering. However, they still rely on particle-based representations of the posterior, which lose the rich information of the full distribution, or tackle a trajectory-level inverse problem that conflicts with the recursive structure of Bayesian filtering. To address this, we propose a new perspective of directly encoding the evolving distribution into flow matching model weights, namely, the Belief Flow Filter (BFF). It is a generative filtering framework that updates model weights via gradient descent at test time to track the posterior evolution. Thereby, BFF bypasses the scalability issue of particle representations or the flexibility limitation of Gaussian assumptions in conventional filters. We theoretically justify that the BFF design is structurally aligned with Bayesian filtering, and its training objective targets the recursive filtering operator. BFF is empirically verified across 5 different physical systems, including ones with chaotic dynamics and highly sparse, non-linear observations. The results show that BFF attains the best score in 8 of 9 metric-benchmark cells across the three standard 1D and 2D PDE benchmarks, and similarly leads on the extreme single-moving-sensor setting and on a real-world-grounded tokamak plasma estimation task, demonstrating its potential to accurately approximate the Bayesian filtering operator in high-dimensional probability space.

Gaussian Splatting has enabled real-time novel view synthesis, but its tightly coupled geometry and appearance representation often require a large number of primitives to reproduce high-frequency texture details, leading to substantial memory and optimization costs. Recent textured 2D Gaussian methods alleviate this limitation by attaching texture maps to Gaussian primitives. However, bridging the fundamental structural gap between discrete Gaussians and continuous 2D grids requires complex parameterizations that introduce severe computational overhead. This overhead fundamentally compromises the original efficiency of Gaussian Splatting, making the balance between detailed texturing and computational agility an unresolved challenge. To address these challenges, we propose LiteTex-GS, a fast and lightweight texturing framework for Gaussian Splatting. Our method initializes an extremely compact representation, assigning minimal local texture to each Gaussian and progressively allocates higher resolution only to primitives with significant reconstruction errors. To maintain a streamlined geometric scaffold, we introduce a contribution- and area-aware pruning strategy that eliminates low-utility Gaussians. Furthermore, to mitigate the gradient dilution caused by texture upsampling, we design a resolution-aware update rule that preserves rapid and stable convergence. Extensive experiments on standard novel view synthesis benchmarks demonstrate that our method achieves competitive or superior rendering quality while using substantially fewer parameters and less training time than existing textured Gaussian baselines.

Repository-level software engineering (SWE) comprises heterogeneous task categories, whose progress under pooled agentic reinforcement learning can be uneven: gains in some categories coincide with regressions in others, while aggregate resolution obscures these changes. Motivated by this category see-saw, we develop a category-aware expert-training and policy-integration framework. Executable task construction and SWE Labeler, an evidence-grounded multi-axis labeling system, organize the training pools. Initial category-specific RL improves average training success while leaving uneven instance-level progress, motivating explicit consolidation of successful behavior and policy-adaptive task selection. Same-origin category experts alternate long-horizon Agentic-miniRL with Refresh-Repair-Expand (RRE): the updated policy refreshes instance mastery, reuses its own verified successful trajectories for Repair SFT, and reselects tasks for further RL. Label-routed multi-teacher on-policy distillation (MOPD) consolidates the experts into one deployable student, with ReLU-gated reward extrapolation keeping only each teacher's improving direction over the reference. Expert training and policy integration require no external model to provide solution trajectories or action targets. We evaluate Pooled RL and Balanced RL, expert development, and single-model integration through aggregate and per-category resolution, the minimum category lift over each joint-RL baseline, and expert-gain recovery. The final MOPD policy achieves mean resolution of 58.04% on Pro-618 and 59.00% on SWE-bench Multilingual, improving over the base model by 5.39 and 2.78 percentage points, respectively.

Long-context language models interface with external knowledge through raw natural language. In retrieval-augmented systems, this creates a persistent index-payload schism: dense vectors enable searchable routing, but models must re-ingest lengthy text payloads for reasoning at O(N^2) attention cost. Existing compression methods further produce private states tied to specific architectures. We introduce Machine-Interpretable Information (MII), the first agent-to-agent (A2A) document-to-state protocol. A dual-timescale state-space Writer compiles documents into a canonical, fixed-bandwidth state (56 tokens), and a lightweight Translator maps it into any frozen Reader's embedding space, reducing query-time cost to O(K). The resulting .mii artifact unifies Retrieval (searchable geometry), Reasoning (global memory), and Reconstruction (grounded details) in a single transferable medium. We demonstrate strong cross-model interoperability across heterogeneous LLMs (e.g., Llama, Qwen, Mistral) -- despite the Writer using a legacy GPT-2 vocabulary, forcing genuine semantic translation rather than token-level memorization. Mechanistic probes reveal modular latent structure: entity representations can be causally traced and zero-shot transplanted between unrelated document states while remaining decodable. To address lexical reconstruction under fixed bandwidth, we propose Residual-MII, a cache hierarchy combining compiled global memory with sparse local evidence. On HotpotQA (7,405 queries), Residual-MII exceeds full-context Exact Match at approximately 7% of the attention FLOPs, suggesting a paradigm shift toward compiled, transferable neural document formats.

Generation-free world action models (WAMs) retain future-video prediction during training but act from internal video features at inference, leaving unclear what these features should preserve for control. Our representation diagnostics show that representations with more predictable future changes need not make linear action decoding easier. Observed future changes provide additional action information beyond the present, and linearly readable action information is spatially concentrated. These findings motivate Action-Relevant Predictive States (ARPS), a compact predictive interface between the video and action experts. ARPS uses a horizon-conditioned state predictor to aggregate intermediate video features into a compact state that supplies all visual context to the action expert. Future-representation supervision trains different parts of this state to predict visual representations at different future times, together with their changes relative to the present. At inference, the supervision branch is removed, and the action expert only uses the learned predictive state computed from current observations. Controlled ablations show that future supervision substantially improves generalization under distribution shift. ARPS achieves 99.2% success on LIBERO and transfers to LIBERO-Plus without adaptation, reaching 87.3% and exceeding Fast-WAM by 39.2 percentage points.

FORM is a domain-specific symbolic manipulation language widely used in particle physics for processing the very large algebraic expressions arising from multi-loop Feynman diagram calculations. Despite its central role in precision theoretical physics, no artificial-intelligence tooling exists, to our knowledge, for assisting physicists in writing FORM code. We show that contemporary large language models (LLMs), including frontier models with hundreds of billions of parameters, achieve a zero-percent execution pass rate on our instruction-following and tutorial-style FORM tasks without documentation in a single attempt, establishing FORM as a genuine zero-shot language for LLMs at the time of writing. We then present a verification-driven data generation pipeline that uses the FORM binary itself as an execution oracle to produce and validate a corpus of 4,633 training examples spanning deterministic computations, open-ended programs, tutorial code, and knowledge question-answer pairs. Fine-tuning a compact open-weights model (Qwen3-8B) with quantized low-rank adaptation (QLoRA) yields a specialist that, evaluated on four complementary benchmarks (840 tasks, single attempt each), decisively outperforms frontier models with up to 756B parameters in execution rate and in strict, FORM-verified output matching on the larger benchmarks, and remains statistically indistinguishable from them on the smaller, harder ones. General reasoning and coding capabilities are preserved within 2.6 percentage points.

Recent advances in large language models (LLMs) have led to the emergence of coding agents capable of performing complex engineering tasks, including register-transfer level (RTL) design and optimization. Existing RTL benchmarks mainly evaluate functional correctness and performance, power, and area (PPA) of the generated RTL designs, leaving agents' ability for \emph{timing closure} under-evaluated. We propose TicTacBench, a benchmark specifically designed to evaluate coding agents' capabilities for RTL-level timing closure under post-place-and-route (post-PnR) evaluation. TicTacBench contains 30 diverse tasks, each provided with a suboptimal RTL design, realistic timing constraints, functional equivalence verification, and timing reports. With over 300 runs of coding agents driven by 8 frontier LLMs, we find that even the best agent can only close 53.3\% of tasks with 7.18\% area-delay product (ADP) degradation and 8.83\% energy-delay-squared product (EDDP) improvement on average. We identify common failure categories that explain why agents fail to close timing. Then we propose TicTacSkill, a new method that guides agents to follow standard timing-closure procedures and improves the Timing Closure Rate by 9\%. These results suggest that while coding agents have made significant progress in RTL design, their timing-closure capability still has substantial room for improvement.

Engineering predictions require physical mechanisms to be translated consistently into equations, discretization, code, and validation, yet errors can propagate despite local checks. Artificial-intelligence (AI) agents automate scientific tasks, but coordinating and independently auditing the theory-to-solver process under physical constraints and human oversight remains unresolved. We introduce a human-in-the-loop, physics-constrained multi-agent workflow where human experts define admissible physics and modeling boundaries, while agents retrieve evidence, derive equations, implement solvers, and audit the theory-to-code chain. Applied to soil-plug evolution during suction-caisson installation, the workflow generated and audited 6 formulations in 2.9 h of agent execution once physical knowledge and inputs were prepared. Among these formulations, adding seepage-driven soil void-ratio evolution to the geometric baseline reduced mean absolute final-heave error from 58.4% to 9.0% across 14 profiles; the selected model further incorporated near-wall dilation and achieved mean absolute percentage errors of 12.4% across 9 final-state cases and 4.2% at the endpoints of 5 process histories. Beyond predictive performance, blinded replay recovered all 9 target problems, while an independent audit uncovered 5 implementation problems after 36 predefined checks had passed. Overall, this work extends multi-agent AI beyond task automation toward human-governed engineering solvers.

Long-term embodied AI will undergo learning, model updates, memory compression, hardware repair, and migration across embodiments. For users who have formed sustained relationships with such systems, these changes raise not only a problem of product consistency but also one of identity continuity: whether the changed system is still experienced as the same particular agent. Existing research suggests that human-AI relationships may develop relational particularity, that robotics and artificial-identity research has identified identity and migration signals across embodiments, and that major updates or platform disruptions can be accompanied by relational loss and restoration desire. This article proposes a user-side framework in which long-term embodied AI is represented through an agent-specific identity representation organized by at least three open identity-content domains: embodied-perceptual, psychological-behavioral, and relational-autobiographical. Information from these domains is not equally weighted; shared history, relational roles, and contingent responsiveness may make some information more identity-diagnostic than others. After system change, users may integrate continuity and discontinuity evidence in a weighted manner, yielding judgments along a continuum from relatively strong identity continuity through ambiguity or partial continuity to clear identity discontinuity. Causal-historical provenance and user participation are treated as contextual evidence rather than a fourth identity-content domain. The framework also proposes identity continuity as a psychological objective for lifecycle design, including memory selection, model updating, and migration across embodiments, under constraints of privacy and user control.

Bimanual interaction produces complementary tactile views of the same physical process, yet existing tactile representation learning largely models the two hands independently or combines them only for downstream prediction, leaving their cross-hand relationship unexplored. To exploit this overlooked structure, we introduce BiView-Touch, a tactile-only framework that completes masked target-hand latents from the remaining visible target-hand regions and the synchronized full contralateral hand. A student encoder with a geometry-conditioned directional decoder predicts full-view EMA latent targets, while temporal and layout counterfactuals encourage sensitivity to synchronized and anatomically organized source information. Controlled ablations and source-context interventions show that BiView-Touch learns structured cross-hand dependence on temporally aligned and anatomically organized contralateral tactile context, rather than benefiting from bilateral input alone. On the public HumanTouch dataset, its frozen representations consistently outperform representative self-supervised baselines across low-label settings. With only 5\% downstream labels, BiView-Touch achieves relative balanced-accuracy gains of 7.1\% on bilateral wrist-motion recognition and 14.1\% on force-derived interaction-phase recognition. We further introduce BVT-20, a 20-task bilateral tactile dataset, and demonstrate transfer across recording sessions and pretraining corpora, including transfer to a held-out bimanual task. Our code and dataset details are available on the anonymous project page: https://anonymous.4open.science/w/biview-touch-review-site-050C/.

Localizing and describing fine-grained differences between near-identical images is a critical yet underexplored capability for multimodal large language models (MLLMs). Existing benchmarks largely assess semantic comparison or single-image grounding in isolation, without jointly requiring faithful description and physical localization. To bridge this gap, we introduce MinCU, a benchmark for grounded minimal-change understanding, where each sample consists of an image pair differing by a single atomic variation in object category, attribute, count, or spatial position, and models are evaluated on their ability to describe the change, localize the changed regions, and identify the changed entity. We further propose Semantic-Guided Implicit Spatial Anchors (SG-ISA), a structured autoregressive method that decomposes prediction into a Think-Locate-Describe sequence. SG-ISA first predicts a semantic cue for the changed concept, then uses discrete spatial anchors as an implicit localization scaffold, and finally generates the change description together with the grounding box. Experiments reveal that even the strongest closed-source MLLMs and recent R1-style reasoning models struggle on MinCU, with most failing to jointly produce accurate descriptions and grounding boxes. Compared to the previous chain-of-thought method, fine-tuning with SG-ISA yields substantial joint improvements in grounding accuracy and description quality while reducing reasoning-token overhead by approximately 26%. These results suggest that an implicit intermediate spatial interface can be more effective than relying solely on model scale for grounded dual-image understanding.

Stochastic reconfiguration (SR) is the standard optimizer for neural quantum states (NQS), but modern NQS often have far more parameters than Monte Carlo samples. We show that in this regime the diagonal shift is more than a numerical stabilizer. It acts as a statistical filter for finite-sample generalization. At a fixed wave function, SR is ridge regression from tangent features to the centered local energy. Its residual is the expressivity gap, the part of imaginary-time evolution outside the current tangent space. This gap is orthogonal to the tangent space in population, but finite batches make it act as noise that SR can overfit. The shift therefore balances shrinkage of useful update directions against variance from fitting sampled residuals. Exact diagnostics on a $4\times4$ Heisenberg graph separate two effects of overparameterization. Larger tangent spaces help when they reduce the expressivity gap, but they can hurt when they overfit a fixed gap. In a $100$-site transverse-field Ising family trained with a foundation NQS, validation risk is U-shaped in the shift while variance decreases, matching the noisy-ridge model. This view leads to multi-shift SR (MS-SR), which averages independent ridge solves at data-adaptive shifts to form a richer, lower-variance spectral filter. Checkpoint-local experiments show that MS-SR lowers validation risk and update variance relative to the fixed-shift SR baseline. We further compare MS-SR and SR in paired online training continuations, with independent endpoint energy evaluations and a separate update-cost benchmark.

Digital direct-to-consumer (DTC) health campaigns are usually measured after the fact. In-flight forecasting commonly relies on a separate classifier for every cutoff and horizon. We treat this task as a dynamic-system problem and build a compact patient world model. The architecture maintains a latent state per patient, learns exposure-conditioned state dynamics jointly with a weekly conversion hazard, and rolls forward into future conversion curves. We evaluate it on a US campaign dataset with 147{,}173 patients and 5.2 million at-risk person-weeks. In a retrospective evaluation conditioned on recorded future exposures, the model forecasts the remaining new-to-brand prescription volume through week 52 with a relative error of 2.9\% from a week-4 cutoff and 0.8--2.6\% from cutoffs at weeks 8--26. The strongest non-recurrent baseline, a pooled-hazard gradient boosting model given the same survival rollout and information, has relative errors of 13.6--33.1\%. Per-horizon classifiers perform substantially worse. A Fisher-information analysis motivates dense next-exposure supervision when conversions are rare. Removing this auxiliary objective increases prescription-volume error by approximately $2$--$14\times$, while providing no consistent disadvantage on the more common specialist-visit outcome. We also evaluate scenario simulation. Switching all future exposure off raises predicted conversion from 0.31 to 0.89, a pattern consistent with selection effects in observational exposure data. This result highlights the limits of interpreting exposure-conditioned rollouts causally.

Heterogeneous integration combines complementary material properties in a single photonic platform. Here, we demonstrate on-chip $f$-2$f$ self-referencing in a heterogeneously integrated Si$_3$N$_4$/LiNbO$_3$ platform, in which supercontinuum generation in a Si$_3$N$_4$ waveguide and second-harmonic generation in a LiNbO$_3$ layer are spatially separated and linked by adiabatic escalator couplers. Pumped by a 1560 nm mode-locked laser with 75 pJ on-chip pulse energy, we detect the carrier-envelope offset frequency with 30 dB signal-to-noise ratio in 300 kHz resolution bandwidth. A comparative measurement confirms that the second harmonic originates in the LiNbO$_3$ layer and indicates opportunity for further improvements through periodic poling. These results establish heterogeneous Si$_3$N$_4$/LiNbO$_3$ integration as a viable building block for self-referenced ultrafast pulse sources in low-loss Si$_3$N$_4$ photonics.

Single-cell foundation models (scGPT, scBERT, Geneformer) achieve cell-type classification accuracy up to 97.5% in our experiments, yet this aggregate accuracy can mask systematic failure on rare, often disease-relevant cell populations that long-tail loss functions are widely assumed to address. We present a systematic benchmark of six long-tail loss functions (cross-entropy, weighted CE, class-balanced loss, focal loss, LDAM, logit-adjusted softmax) across three architectures and three datasets (Multiple Sclerosis, Zheng68K, human Pancreas), totaling 162 controlled training runs (3 backbones x 3 datasets x 6 losses x 3 seeds). The gap between overall accuracy, Macro-F1, and rare-class recall under plain cross-entropy is consistent across all nine (architecture, dataset) settings, driven by dataset structure rather than pretraining. Rare-class failure itself splits into two regimes with distinct embedding-geometry signatures, visible before any loss is chosen: some classes are recoverable by the right loss, while others retain linear separability yet are absorbed into unrelated classes' neighborhoods under every evaluated loss and architecture. Among the recoverable classes, the efficacy of reweighting is predicted by a class's absolute training-set size, rather than its share of the dataset or the dataset's overall imbalance ratio. Class-balanced loss and LDAM are the most consistent choices across all nine settings, while logit adjustment trades rare-class precision for recall rather than improving both. Our results give both a reusable benchmark and mechanism-grounded practical guidelines for combining foundation models with imbalanced biological data.

Low-rank adaptation (LoRA) has become a key technology for serving large-scale personalized large language models and diffusion models in the cloud. However, the co-occurrence patterns, resource contention relationships, and evolutionary regularities of adapters under production inference workloads have not been systematically or quantitatively studied. Based on GenTD26, Alibaba's production diffusion model inference dataset, this paper adopts a graph-theoretic framework to construct an adapter co-occurrence network and conducts a characterization from both static structure and dynamic evolution. Our main findings are as follows. (1) The co-occurrence network is extremely sparse, and adapter usage frequency follows a significant heavy-tailed distribution. (2) Introducing the first adapter incurs a 66.1% execution-latency overhead, with diminishing marginal costs afterwards. (3) Co-occurrence relationships are driven by base models: in 90.6% of multi-adapter requests, all adapters share the same dominant base model; 66.2% of significant co-occurrence edges connect same-model adapter pairs; and in 85.8% of multi-adapter requests, all adapter pairs form significant co-occurrence edges. (4) The adapter ecosystem exhibits a core-periphery bipolar structure, with a weekly Jaccard similarity of 0.696 at the model level and a churn rate of 54.5% for the top-10 hottest models within a 12-hour window. Based on these findings, we propose a preloading strategy built on top-k co-occurrence statistics; offline experiments show that it covers 81.0% of test-set co-occurrence pairs at k=3, and sensitivity analyses across frequency thresholds and time windows verify the robustness of the conclusions. These results provide a data-driven basis for cache preloading, adaptive scheduling, and GPU memory management in LoRA inference services.

Social bot detection is essential for protecting online platforms from misinformation amplification, coordinated manipulation, and distorted public discourse. However, large language models have made social bots much harder to detect from text alone because semantic camouflage is now cheap, fluent, and scalable. The resulting challenge is modality conflict: an account may look human-like in semantics while remaining suspicious in graph structure, profile attributes, or cross-modal consistency. Recent graph-based detectors tackle this limitation by adding graph-side complexity, such as sparse prototype selection, adaptive gating, or architecture-specific control logic, yet our experiments suggest that complexity alone is not the most reliable way to resolve such conflict. We therefore propose CSC, a calibrated-simplicity framework for conflict-aware LLM-era social bot detection. The framework combines three design choices: a simplified prototype-guided graph expert that retains useful structural biases while removing unstable graph-side heuristics, calibrated simplex-constrained fusion that aligns heterogeneous confidence spaces before late fusion, and a lightweight inconsistency expert that models cross-modal disagreement. Experiments on TwiBot-22, TwiBot-20, and MGStBot-large show that \textsc{CSC} improves calibrated operating-point decision quality while remaining competitive across external benchmarks. Further analyses show that calibration improves confidence reliability, the inconsistency expert mainly provides localized corrections in high-conflict or near-threshold regions, and simplified graph-side control yields a better stability-cost trade-off. A targeted semantic-camouflage stress test further shows that replacing selected bot text with matched human text sharply degrades the standalone text expert while leaving graph and fused evidence stable on a balanced challenge set.

Graph databases are frequently positioned as categorically necessary for connected-data workloads, yet the systems dimension along which they actually differ - query planning, indexing, and data-readiness cost - is rarely isolated from vendor framing. We construct a synthetic, biomedical-shaped property graph (1.02 million nodes, 5.34 million total node and edge rows) and a twenty-query workload spanning neighborhood lookups, bounded paths, set intersections, anti-joins, grouped aggregation, top-k ranking, temporal filters, full scans, and relational joins. We benchmark Corvic AI - a purpose-built columnar query engine underlying Corvic's ontology management layer ("memories")- against seven purpose-built or graph-extension database systems (LoraDB, Ladybug, DuckPGQ, Memgraph, Neo4j, HugeGraph, and FalkorDB) at three graph scales spanning three orders of magnitude. We report query latency geomeans, bulk-ingest throughput, point-update latency, and answer correctness for each system, and we derive a simple total-cost-of-ownership model that expresses the ingest/query trade-off as a function of query volume. Our central finding is that no system in this sample is categorically fastest: a native graph engine (Ladybug) outperforms Corvic AI on narrow, bounded-neighborhood shapes, while Corvic AI is faster on shapes that scan or join a large fraction of the graph, and a system implementing graph query syntax via SQL/PGQ (DuckPGQ) is measurably slower purely due to query-plan choice. The dominant cost differential in our data is not query latency but the cost of making data queryable at all: bulk-ingest throughput varies by three orders of magnitude across engines (5.0k-4.3M rows/s), a gap that a simple crossover-point calculation shows dominates total cost for any workload with fewer than roughly 105 queries per data refresh.

Modern LLMs with QK-normalization, gated attention, learned attention sinks, or logit softcapping exhibit weaker persistent attention sinks, on which existing KV cache eviction methods primarily rely. We observe that across these models, weaker sinks co-occur with greater value-vector dispersion relative to key-vector dispersion. Motivated by this value-side dispersion, we present ValueDiff, a value-geometric eviction that ranks tokens by the L2 deviation of their value vectors from the cache mean. The same score arises as the minimal-disturbance eviction under a max-entropy assumption about future attention. We evaluate under fixed cache budgets, with eviction at every block boundary during prefill and at every decoding step during generation. On RULER at a tight 2k token budget, ValueDiff retains 88--99\% of dense across seven sink-suppressed models (best on 6 out of 7). On LongBench at the 4k budget, ValueDiff averages 92\% retention across sink-suppressed models versus 83\% for the strongest prior baseline. On MATH-500, ValueDiff is the strongest non-dense method on every sink-suppressed model tested at the 25\% cache budget, outperforming prior methods by up to $\sim$20 points on gated-attention models. Across all three benchmarks, value geometry emerges as the more reliable query-invariant eviction signal for sink-suppressed models.

In environments with large movable obstacles, detour-only navigation can be inefficient or even infeasible, while obstacle interaction requires reasoning about navigation benefit, feasible placement, and executable manipulation. We present a hierarchical navigation among movable obstacles (NAMO) framework for mobile manipulators. At the high level, the planner identifies key blocking obstacles from reference paths and searches for relocation plans that jointly satisfy geometric, manipulation, and downstream navigation constraints. When direct relocation is hindered by other movable objects, a large language model (LLM) is selectively invoked to infer auxiliary manipulation dependencies, which are then verified by deterministic geometric planning. To execute the resulting relocation goals, we define discrete contact modes on the surfaces of box-shaped obstacles and select contact faces and regions online based on position and orientation errors, enabling straight, side, and corner pushing through contact switching. A recurrent reinforcement-learning policy coordinates the mobile base and manipulator to track tool center point (TCP) targets while preserving end-effector reachability during sustained pushing. Simulation and real-robot experiments demonstrate feasible navigation-manipulation in detour, single- and multi-obstacle relocation, and dependency-constrained scenarios, validating the framework for interactive navigation with large non-graspable obstacles. The open-source project is available at https://cloudytosunny.github.io/NAMO_DCPushing/.

This paper presents a comparative evaluation of dense embedding models for semantic candidate-job matching in high-volume staffing workflows. Incoming job descriptions are converted into structured English search text and language-specific keywords through LLM-based parsing, and candidate profiles are indexed as semantically enriched resume representations. We evaluate EmbeddingGemma (base) against EmbeddingGemma fine-tuned with Cached Multiple Negatives Ranking Loss (MNRL) within a unified hybrid retrieval pipeline that fuses vector similarity and full-text relevance via reciprocal rank fusion (RRF), and benchmark both against the MPNet model on a batch comparative evaluation dataset scored through the deployed job-candidate matching scoring pipeline. We further document, with mathematical detail, the broader set of contrastive fine-tuning objectives considered during model development (including AnglE/CoSENT-style refinement) and the empirical rationale for retaining Cached-MNRL-only adaptation as the preferred configuration. To support reproducible model selection, we define a broader evaluation framework comprising standard information retrieval metrics (Recall@K, mean reciprocal rank, nDCG) under the exact hybrid-retrieval protocol; the metrics used for the evaluation reported in this paper are fine-tuning convergence diagnostics and a batch comparative evaluation using the deployed AI-Match score and an independent LLM-as-a-Judge relevance score, and we state this scope explicitly rather than implying the full framework was measured. The paper addresses the gap between general-purpose embedding benchmarks and enterprise job-candidate matching constraints, providing a structured basis for comparing embedding strategies under realistic job-candidate retrieval conditions.

Manipulation under time constraints requires both accurate actions and an execution rhythm that matches the evolving scene. This becomes critical when a robot must intercept moving objects or complete a sequence of adjustments before a deadline. Although one-step policies reduce generation cost, their directly predicted action sequences leave temporal allocation implicit. We propose Shared Execution-Clock Drifting (SECD), which makes execution rhythm an explicit part of one-step action generation. Conditioned on an observation and a latent sample, the policy jointly predicts a progress-indexed action curve and a shared monotone clock that maps fixed control times to locations on the curve. Demonstration-derived alignment anchors this decomposition, which is trained jointly through drifting on the decoded actions. The resulting policy retains a fixed-rate control interface and requires one network evaluation. We evaluate SECD across four real-robot tasks with inference on NVIDIA Thor. Across 300 trials, it achieves 77.00% task-averaged success and outperforms the evaluated one-step baselines on every task, including 91% success in cup retrieval from a 16 m/min conveyor and 54% in restoring and folding a crumpled shirt within 90 s. A fixed-clock variant reaches 79% on the same conveyor protocol. Complementary state-based RoboMimic experiments, including cross-seed ablations on Transport and Square, further support the joint design of the temporal representation and demonstration alignment. Project page: https://secd-anonymous-ewn.pages.dev/

Despite the remarkable progress of LLM-based and knowledge graph-augmented Radiology Report Generation (RRG) methods, existing techniques still suffer from inherent defects. Conventional LLM-only models lack structured medical prior knowledge, resulting in frequent medical hallucinations and low diagnostic interpretability. Current knowledge graph-enhanced schemes adopt static one-round knowledge fusion with single-source knowledge, incapable of dynamic knowledge updating according to generation feedback. This paper proposes a novel Multi-Agent Collaborative iterative framework for X-ray Radiology Report Generation, termed MAC-RRG. Inspired by multi-agent technology, our framework constructs a closed-loop optimization paradigm based on task decoupling and collaborative reasoning. Specifically, the framework first generates a preliminary radiology report from input X-ray images via a vision encoder and a basic LLM. Subsequently, a multimodal knowledge graph (MM-KG) agent mines structured disease correlation and anatomical knowledge from medical knowledge graphs, while an auxiliary knowledge agent extracts unstructured domain knowledge from public medical databases. The multi-source knowledge acquired by dual agents is fused and embedded to guide the LLM in iteratively refining the initial report. Extensive quantitative and qualitative experiments on mainstream X-ray RRG datasets, including IU X-ray, MIMIC, and CheXpert Plus, fully verify the superiority of our proposed method. The source code and pre-trained models have been released on https://github.com/Event-AHU/Medical_Image_Analysis

3D reconstruction from a lengthy video stream input poses a dilemma for feed-forward reconstruction models (FFRMs), that a whole-stream inference context cannot be retained under limited GPU memory.Recent studies seek to resolve this problem via a trade-off between the integrity of inference context and GPU memory usage, which either suffer from a rapid memory inflation or degraded context integrity due to artificially capping memory usage.Driven by our key observation that the initial saliency of a token reliably dictates its long-term importance across the stream, we propose RegVGGT, a training-free token regulation method which aggressively regulates the tokens of incoming frames.By admitting at most 1% of tokens per frame to update the context memory, our method dramatically suppresses memory inflation as the stream progresses.Equipped with a FlashAttention-compatible token saliency estimation scheme, RegVGGT is capable of processing thousands of frames on a consumer-grade GPU with negligible compromise to reconstruction quality.Extensive experiments demonstrate that RegVGGT achieves state-of-the-art performance on long-horizon benchmarks across diverse FFRM prediction tasks, surpassing prior FFRM-based stream reconstruction baselines by a large margin.

Vision-language-action (VLA) policies increasingly incorporate structured intermediate supervision beyond action labels. Yet specifying what an intermediate representation should encode leaves open how action prediction learns to depend on it. We introduce \textbf{SCULPT-VLA}, a policy that learns structured control through staged action grounding. Its action-conditioning state comprises complementary factors for task progression, scene dynamics, and spatial grounding. Training first forms these factors with teacher scaffolds, then grounds coarse action prediction through their composition as scaffold inputs are withdrawn. Direct perceptual access is subsequently restored for continuous refinement, combining the learned state with perceptual detail. The curriculum separates learning to condition actions on structure from refining continuous control. Deployment requires neither teachers nor discrete-action autoregression. SCULPT-VLA achieves higher average success than shared-backbone baselines on LIBERO, SimplerEnv-WidowX, and RoboTwin 2.0 Full. On SimplerEnv-WidowX, final success is 83.5\%, versus 71.3\% when Stage-II action learning directly accesses vision and language. Across four physical robot tasks, average success under the tested distribution shifts reaches 58.1\%, compared with 45.6\% for $π_{0.5}$. Training ablations and factor-wise interventions support the staged design and show that the learned state continues to contribute to control after direct perceptual access is restored.

In a decentralized multi-robot team under partial observability, the fact that decides a robot's next action is often visible only to a teammate. Existing decentralized methods communicate kinematic information, such as position or planned trajectory, which cannot convey what the teammate perceives. Learned communication in multi-agent reinforcement learning (MARL) can carry perceptual content, but the resulting messages are task-coupled and opaque. We propose Latent Telepathy. Each robot broadcasts the perceptual latent vector it already computes for its own use, the output of an encoder trained with a self-supervised joint-embedding predictive objective, frozen, and shared across the team. A teammate learns to act on it from task reward alone. Because the encoder already runs for perception, the message costs no additional computation and a single compact vector of bandwidth. Because the encoder is frozen before any policy is trained, the message means the same thing to every robot, and the receiving robot is never told what it means. We evaluate Latent Telepathy with a content-controlled protocol in which bandwidth, latency, topology and receiver are held fixed and only the message content varies. Broadcasting the latent lets a navigator avoid an occluded hazard in 99.7% of episodes, matching a noiseless hand-designed message. Position and trajectory messages remain at chance, and the raw camera image, 186 times wider, is less reliable than the compressed latent. The result holds from a discrete gridworld to rendered pixels under continuous velocity control, and the encoder decodes the hazard from a physical robot's camera in 102 of 102 live decisions. We also identify a requirement for porting MARL communication results to continuous control, that the decision a message informs must remain reachable by exploration, and show how to restore it.

Peer-review evaluation is increasingly being automated with LLM-as-a-judge metrics, but this creates a measurement risk. A review may receive a high score because it is fluent, organized, and polished, rather than because it provides a strong evaluation of the paper. This risk is especially important in AI-assisted reviewing, where reviewers may use LLMs to improve clarity or presentation while preserving the underlying judgments. We propose a statistical framework for testing whether peer-review evaluation metrics capture substantive review quality beyond surface-level linguistic form. The framework compares original human reviews with faithful LLM rewrites that preserve the same evaluative content while changing wording and presentation. Using a dataset comprising 4,044 meaning-preserving rewrites derived from 674 human reviews from ICLR and NeurIPS, we evaluate 29 content-oriented peer-review evaluation metrics drawn from four prior works through complementary tests of surface sensitivity and robustness. Although these metrics are intended to capture review properties beyond surface-level, writing-dependent characteristics, we find that sensitivity to rewriting is widespread. Under our primary analysis, 23 metrics assign significantly different scores to reviews whose evaluative content is preserved, while only six satisfy our robustness criterion. The patterns are largely consistent across two LLM judge models, suggesting that the issue is not specific to a single judge. These findings show that many peer-review evaluation metrics partially conflate review quality with linguistic presentation, and indicate that robustness to meaning-preserving rewriting should be validated before such metrics are used to compare human-written, AI-assisted, and AI-generated reviews.

Temporal logic is a formal language for reasoning about system behaviors over time. Signal temporal logic (STL), in particular, has been used to encode spatio-temporal requirements for control synthesis in multi-agent systems, often under the assumption that agents are cooperative and their dynamics are known. However, real-world multi-agent applications, such as autonomous driving, typically involve stochastic and uncontrollable agents. Recent work explored robust control with worst-case or probabilistic formulations, but remains limited in that it either (1) certifies strict satisfaction of STL constraints without addressing feasibility recovery, or (2) relaxes infeasible constraints with ego-centric objectives. In this paper, we propose a model predictive control (MPC) framework that treats feasibility repair as a Pareto optimization problem to explicitly characterize tradeoffs among agent objectives. We further provide a probabilistic certificate on STL violation rate to formally quantify uncertainty under stochastic and uncontrollable agents. The proposed framework is evaluated on two autonomous driving scenarios. Results show that the framework recovers feasible control with demonstrated safe behaviors.

Time series forecasting underpins critical decision-making across diverse domains. While large language models (LLMs) offer promising reasoning capabilities, existing LLM-based time series forecasting approaches either reduce them to numerical predictors that bypass their strengths, or allow direct forecast generation that destabilizes predictions in non-stationary settings. We introduce CTRL, a framework that decouples semantic reasoning from quantitative prediction. A frozen backbone generates base forecasts, while specialized LLM agents function as controllers that analyze backbone prediction errors through decomposed trend, seasonal, and irregular components, grounding reasoning in interpretable temporal structure. Each agent outputs compact control signals that a lightweight residual decoder translates into forecast corrections. CTRL incorporates label-free test-time adaptation that detects distribution shift from input statistics alone and readapts control signals with only 3-24 LLM calls via caching. CTRL is explicitly designed to improve robustness under non-stationary temporal dynamics and distribution shift, while remaining competitive on highly stationary time series where adaptive correction provides limited additional benefit.

Advanced Air Mobility (AAM) systems require scalable coordination mechanisms to manage large fleets of aerial vehicles operating in shared, capacity-limited airspace. In such environments, different operators may have private preferences over trajectory characteristics, such as travel time, fuel consumption, or deviation from nominal routes. If centralized traffic management relies on self-reported preferences, operators may strategically misreport their costs to obtain more favorable trajectories. This paper proposes a multistage motion planning framework augmented with mechanism design to enable collision avoidance for AAM systems with privately known costs. The proposed approach integrates convex safe corridor construction with a VCG-inspired mechanism to ensure conflict-free passage through constrained airspace while incentivizing truthful revelation of private preferences. Simulation results demonstrate safe and decentralized coordination among agents with heterogeneous preferences.