OrangeBot.AI Digest — 2026-08-04

The MCP server for marketing data

The AI SDR that runs outbound for you

Japanese for the life you're actually living in Japan

Turn your Mac into a spinning vinyl player.

A terminal built for Claude Code w/ one-click session resume

Your voice-first AI assistant for macOS

Duolingo for learning anything

Create and share beautiful guides from places you love

A proactive AI executive assistant for founders

Build, ship, and improve at scale

Visualize your revenue on a 3D globe

Make videos and post them everywhere with just one tool

Build and customize your own calendar agent you can text

One memory for every AI you use

The trusted AI investment agent, from insight to action

Speech and audio generation is often needed in animation dubbing, audio drama, movies, advertising, games, podcasts, and short-video production. In these scenarios, creators may need to design voices without reference recordings, control speaker styles with natural language, support acoustic scenes with environments and audio effects, and later reuse the designed voices. Therefore, it is important to support multi-speaker speech and audio generation for both instruct and zero-shot tasks. The instruct task requires a caption of the environment, speaker styles, and fine-grained content, while the zero-shot task uses reference audio together with the same fine-grained content. We address these tasks from both the data and model sides. First, we propose SwanData-Caption, which cleans raw speech and audio data, adds targeted synthetic coverage, and annotates diverse and accurate multi-level captions. Then, we propose SwanTale, a multi-speaker expressive speech and audio generation model that supports both zero-shot and instruct tasks. We introduce SwanVAE to support high-quality multi-audio-modality generation. Then, we adopt reward-conditioned quality control and Engram conditioning, along with Unified MoE for multi-task and multi-audio-modality modeling. In addition, we use curriculum learning and GRPO post-training to let the model progressively learn and strengthen its capabilities. Experimental results show that SwanTale leads on multiple key zero-shot and instruct metrics, achieves the best expressiveness scores in both tasks, and supports complex instruct generation involving multi-speaker speech and audio. Demos can be found at https://swanaigc.github.io/\#swantale.

Large language model (LLM) agents increasingly undertake long-horizon tasks that require sustained reasoning, tool use, and revision across many interdependent steps. However, existing agent harnesses maintain task execution, task state, and completion assessment within a growing context, making the state difficult to track and allowing incorrect self-assessments to propagate into later decisions. We reformulate long-horizon execution as a task-state management problem and propose LongHorizon-Harness, which maintains the task state explicitly outside execution and updates it only with facts independently verified from the environment. Its Manage-Execute-Audit(MEA) loop uses a manager to maintain the task state and determine the next subtask, a fresh-context executor to perform it, and a read-only auditor to verify the resulting environment state before the next round. A lightweight AgentAdapter supports interchangeable model and harness backends without modifying their native agent loops. LongHorizon-Harness improves Qwen~3.7-Plus from 51.8% to 80.7% on WeaveBench, from 69.7% to 77.2% on Terminal-Bench~2.1, and from 2.8% to 8.3% on OSWorld~2.0. It also raises Claude Opus~4.7 from 20.0% to 34.3% on an OSWorld2.0 subset, demonstrating consistent gains across models, harnesses, and interaction domains.

On-policy (self) distillation (OPSD) is increasingly adopted for language-model post-training. It strengthens the teacher with privileged information but can induce a privilege illusion: the student learns privilege-dependent behavior it cannot reproduce from its inference-time context, yet behaves as if the training-time privileged information remained available, ultimately degrading performance. In this paper, we identify information asymmetry between the privileged teacher and the student at inference as the root cause of this failure in OPSD. To resolve this asymmetry, we propose Dual-Anchored Policy Distillation (DAPD), a unified framework with two levels of anchoring. Dual-Path Anchoring (DPA) introduces a self-conditioned bridge and aligns reference and rollout behavior along two matched-information paths, preventing privilege-dependent behavior from being transferred to the inference-time student. Dual-Source Anchoring (DSA) applies these paths in both reference-to-rollout and rollout-to-reference directions, reducing reliance on privileged reference guidance while preserving correctness supervision. Extensive experiments show that DAPD significantly alleviates privilege illusion, outperforming OPSD on Qwen3-4B by +2.00 points on average across tasks. Notably, its gains persist across scales, reaching +2.69 at 4B and +2.78 at 32B.

Existing skill generation methods largely rely on heuristics or pipeline-style consolidation, which must be specially designed for different evidence sources. In contrast, learning-based approaches offer a more unified way to model skill generation across heterogeneous sources. However, learning-based skill generation remains challenging because skills lack a natural supervision signal based on relevance or correctness; their value can largely be determined only by whether they improve the behavior of the agent on downstream tasks. To address this challenge, we propose Skill-α, a reinforcement learning method for progressively generating high-quality agent skills. Specifically, we formulate skill generation as a sequential editing process that decomposes skill construction into individually evaluable edits, and introduce a novel rollback reward that evaluates each edit by comparing downstream execution under the original and edited skills on an anchored query. Extensive experiments show that Skill-α generates more effective skills than methods based on heuristics or pipelines in both document-to-skill and experience-to-skill settings. Under the main GPT-4o worker, Skill-α improves average downstream success rates over the strongest skill-generation baseline by 3.3 points on CL-Bench and 6.7 points on tau2-bench. Further ablations validate the importance of rollback reward and progressive generation.

Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-mixed, combining visual signals with linguistic priors and teacher-specific effects. The key challenge is to estimate which corrections are supported by visual evidence, not merely where or how strongly to distill. We introduce Visual Attribution Distillation (VAD), a counterfactual target-reconstruction algorithm that estimates the visually attributable part of a teacher correction. At each student-generated prefix, VAD evaluates the same fixed teacher with the relevant evidence present and removed. The corresponding change in centered log-probabilities defines ut, a signed proxy for the visual evidence direction that estimates how revealing the evidence supports or refutes candidate tokens. VAD projects the original correction onto this proxy to obtain an intervention-aligned component and a proxy-unexplained residual, then reconstructs a student-anchored target from the former. During training, this reconstructed target supplies the primary supervision signal, while the privileged teacher contributes a weak regularizer. Across six fine-grained visual benchmarks at 4B and 9B scales, VAD outperforms direct privileged-view distillation and visual-advantage weighting. Token- level and controlled-target analyses show that the proxy-aligned component is enriched in task-relevant visual corrections and yields stronger target shifts, especially when evidence refutes a mistaken answer. These results support counterfactual target reconstruction as an effective alternative to source-mixed supervision.

Sparse retrieval underpins modern search systems, from web search to retrieval-augmented generation. Existing work has introduced Learned Sparse Retrieval (LSR) to push beyond exact lexical matching toward richer semantics. Yet LSR has so far remained tied to encoder-style bidirectional architectures, and its extension to multimodal settings still relies heavily on auxiliary cross-modal modules. To address these limitations, we introduce UEmbed (Unified Embedding), a decoder-only multimodal embedding model that produces both sparse lexical and dense representations in one causal forward pass. UEmbed appends N learnable special tokens to the input and partitions the vocabulary into N disjoint subsets. Each token's causal hidden state predicts sparse weights over its assigned subset, and the N subsets are concatenated into the full sparse vector. Trained on public data, we release UEmbed at 2B, 4B, and 9B scales. UEmbed-9B reaches 71.8 (dense) and 71.0 (sparse) on MMEB-v2, outperforming multimodal embedding models trained on publicly available data (e.g., RzenEmbed). On BEIR, UEmbed also remains competitive with strong dense and sparse baselines. Furthermore, we demonstrate the practical utility of UEmbed across three dimensions: effectiveness, efficiency, and agentic applications. Overall, UEmbed offers a new paradigm: it unifies dense and sparse embeddings in one model, while further extending sparse retrieval to unify text and multimodal inputs.

Computer-Aided Design (CAD) underpins modern engineering, yet converting existing shapes into editable models still demands substantial expert effort. Most AI systems emit the entire CAD program in a single pass, never inspecting the intermediate geometry. In contrast, human engineers build a part feature by feature, checking after each operation what remains to be modeled. We introduce CADENA (Spanish for "chain"), a model that reconstructs a 3D mesh as a parametric CAD program, growing its sequence of operations one at a time and comparing the target with the currently predicted geometry at every step. We also address the lack of benchmarks for evaluating reverse-engineering methods on mechanical parts, introducing CADENA-Bench, a benchmark that measures performance across categories of mechanical parts. CADENA outperforms prior methods on CADENA-Bench and on the DeepCAD, Fusion 360, and MCB datasets. Code is available at https://github.com/zhemdi/cadena, model weights at https://huggingface.co/kulibinai/cadena, and CADENA-Bench at https://huggingface.co/datasets/kulibinai/cadena-bench.

Controllable video generation models are increasingly being developed as world models. Accordingly, evaluating them in this role extends beyond the apparent appearance of generated videos to the inherent reactivity of the worlds they depict: the ability to infer from the scene state how the world should react and to generate plausible consequences not explicitly described in the input. Yet existing benchmarks mainly assess visual quality or explicit instruction fulfillment by checking whether requested actions and interaction outcomes are realized, leaving inherent reactivity underexamined. We introduce WorldExam, a hierarchical diagnostic benchmark spanning four levels: Visual Quality, Control Adherence, Spatial Consistency, and World Reactivity. It comprises 1,474 cases across eight dedicated tasks and supports unified evaluation of camera-, action-, and language-driven model paradigms. The World Reactivity level evaluates scene-conditioned reactions and goal-directed behaviors beyond what is explicitly specified in the input. Evaluation of 20 representative models reveals a clear capability split. Camera-driven models excel at camera control, but their interfaces do not support dynamic interaction; action-driven models control subjects more precisely but often leave the world unresponsive; and language-driven models perform better on interaction but follow complex controls less faithfully. No model combines broad task coverage with consistently strong performance, showing that high visual quality and explicit instruction fulfillment do not guarantee inherent reactivity.

Recent Vision-Language-Action (VLA) models for autonomous driving (AD) increasingly utilize chain-of-thought (CoT) supervision to enhance the reasoning capabilities of their Vision-Language Model (VLM) components, yet existing annotation pipelines commonly expose the teacher model to the logged ground-truth (GT) future trajectory. We empirically show that this induces trajectory anchoring bias: teacher models rationalize the revealed outcome rather than infer a decision from scene evidence, producing less causally faithful CoTs and substantially more severe hallucinations, especially in causally challenging scenes. Removing the GT trajectory eliminates this shortcut, but open-ended trajectory generation entangles high-level decision-making with precise geometric synthesis and low-level dynamics. To make trajectory-level driving decisions verifiable without requiring open-ended trajectory synthesis, we introduce Autonomous-Driving Multiple-Choice Question (AD-MCQ), which casts planning as selection among explicit trajectory candidates. Taking this a step further, we propose Deferred Exposure of Future Trajectories for RLVR (DEFT-RLVR) to transform future trajectories from pre-decision anchors into post-decision verification targets. Experimental results show that DEFT-RLVR improves AD reasoning while preserving or even enhancing general visual capabilities. With VLM-only inference and controllable difficulty through candidate construction, AD-MCQ provides a flexible, scalable, and extensible foundation for future research on verifiable AD reasoning.

Agent skills have become an important mechanism for equipping language-model agents with reusable procedural knowledge. However, providing skills alone does not guarantee that current models can effectively identify, apply, and coordinate them. To improve skill-use capabilities, we introduce SKT, a verified data synthesis pipeline that constructs skill-grounded tasks and executable trajectories from large collections of agent skills. SKT selects suitable single-skill and multi-skill configurations, synthesizes tasks through rule-based and agent-based verification with feedback-guided repair, and retains only successful trajectories that substantially use every required skill. Using 2,000 public skills, SKT produces 4,000 task packages and 27,164 verified trajectories. Based on the same pipeline and a disjoint test pool, we further construct SkillEval, a held-out executable benchmark for evaluating skill use. Experiments across diverse models, benchmarks, and agent harnesses show that supervised fine-tuning on SKT-generated trajectories consistently improves skill-use performance. Verification ablations, cross-harness evaluation, and scaling experiments further demonstrate that these gains depend on high-quality supervision, extend beyond a single agent interface, and increase with broader skill coverage. Together, these results establish verified data synthesis as an effective and scalable approach for skill-use training.

We introduce DiffusionGemma, an experimental open-weight language model that uses discrete diffusion to generate text at exceptionally high speed. Rather than decoding one token at a time, DiffusionGemma iteratively refines blocks of 256 tokens in parallel, avoiding the sequential decoding bottleneck of conventional autoregressive (AR) large language models. Instead of training from scratch, we obtain DiffusionGemma by fine-tuning the mixture-of-experts Gemma 4 model with 3.8B activated and 25.2B total parameters. Our compute-efficient two-stage training pipeline uses fewer than 10% of the starting AR model's total training token budget. The first stage uses supervised fine-tuning to teach bidirectional denoising, while the second stage combines reinforcement learning with sampler distillation to jointly improve generation quality and inference efficiency. DiffusionGemma establishes a new Pareto frontier for the trade-off between generation speed and model capability. Averaged across our full evaluation suite, it generates around 20 tokens per forward pass and achieves roughly 1,500 output tokens per second on a single NVIDIA H100 GPU, which is substantially faster than AR models even with state-of-the-art speculative decoding. DiffusionGemma also retains the starting model's support for thinking mode, multimodal inputs, and long contexts. Despite diffusion fine-tuning, it remains capable of AR generation with only minor performance degradation, suggesting a path toward hybrid diffusion-AR decoding.

Real-world software development requires coding agents to operate in shared workspaces where users may inspect and modify code during an ongoing task, yet existing repository-level benchmarks typically evaluate agents working alone or restrict user participation to messages. This leads us to ask: how do coding agents understand and respond to code changes in a shared workspace? We introduce SWE-Touch, a framework that stress-tests this setting through validated Counter-Edits: plausible edits to task-relevant code that conflict with task completion. SWE-Touch mines task-critical regions from multiple repair trajectories, uses a separate User Patch Generator to construct the edits, and injects them with contextual user messages when agents reach the relevant code. We evaluate nine coding models on SWE-bench Verified, with additional experiments on longer-horizon tasks from SWE-Bench Pro and DeepSWE. Counter-Edit lowers average resolve rate by 7.7 percentage points on SWE-bench Verified, with degradation also persisting on both longer-horizon benchmarks. Trajectory analysis links these failures to limited awareness of the evolving workspace: agents may retain conflicting code or replace it without sufficiently re-inspecting the repository and validating the revised code with targeted tests. These findings show that strong autonomous performance does not yet ensure the state awareness and adaptive behavior needed for shared-workspace collaboration, and point to detecting workspace changes, reconciling conflicting edits with the task, and verifying the affected behavior as key capabilities for future optimization.

Reinforcement learning (RL) post-training of Vision-Language-Action (VLA) models has shown strong promise for robotic manipulation. Among RL methods, critic-based approaches rely on a value estimator that predominantly operates on single-frame observations or single-frame VLM backbone latents, which is a fundamental mismatch with the partially observable nature of robot control. A naive approach to incorporate observation history into the critic incurs exponential complexity with high-dimensional visual space, and still fails because pure scalar-return regression provides insufficient supervision for learning cross-temporal dynamics. We identify the root cause as a state approximation problem: without an explicit world modeling objective, the critic's representation cannot capture the temporal structure needed for accurate value estimation. To address this, we propose the World Critic Model (WCM), built on a lightweight LeJEPA architecture; WCM jointly predicts future latent state and estimates values, such that the critic's representation is explicitly trained to capture temporal dynamics rather than merely regress scalar returns. WCM integrates seamlessly into both on-policy and off-policy training pipelines and is compatible with state-of-the-art VLA backbones including Pi0, Pi0.5, and OpenVLA-OFT. Extensive experiments on 149 tasks across four benchmarks demonstrate that WCM consistently achieves state-of-the-art performance in both in-distribution and out-of-distribution settings, with particularly strong generalization gains. We further validate WCM on seven real-world manipulation tasks using OpenVLA-OFT and Pi0.5 with off-policy RL, confirming stable deployment across diverse settings.

Optimization-based latent reasoning improves large language model outputs by optimizing instance-specific continuous states at test time while keeping model parameters frozen. Existing methods, however, typically connect these states to the reasoning trajectory through decoded tokens, making sequence-level credit assignment indirect and obscuring how latent updates shape subsequent reasoning. We introduce GradCuit (gradient through circuit), which inserts optimizable latent states at a selected Transformer layer between the hidden representations of the prompt and the generated continuation. Causal self-attention provides every continuation-token log-probability with a differentiable path to every preceding latent state through the remaining Transformer blocks, enabling reward-weighted gradients from the entire continuation to be assigned directly to the latents. Across five instruction-tuned backbones, three reasoning benchmarks, and two answer formats, GradCuit achieves an average accuracy of 64.5%, outperforming chain-of-thought prompting by 6.6 percentage points and the strongest competing method by 2.4 points. GradCuit also demonstrates greater robustness: across seven learning-rate settings, it consistently outperforms LatentSeek while reducing the standard deviation of accuracy from 1.53 to 0.82, and even its random-walk variant remains competitive with LatentSeek. For interpretability, token-level gradient attribution reveals that latent influence concentrates on reasoning-connector tokens, while layer analysis identifies early-to-middle Transformer layers as the most effective optimization space. By directly optimizing internal reasoning from outcome feedback, GradCuit opens a new axis of robust and interpretable test-time scaling, where LLMs adapt how they reason rather than merely regenerate, sample, or rerank outputs.

Video motion transfer aims to animate a target object using dynamics from a reference video. Existing formulations largely rely on fixed structural correspondence, which becomes ill-defined when reference and target objects differ substantially in morphology, articulation, or deformation mechanisms. We introduce Motion Beyond Morphology, a perspective that seeks to transfer motion beyond fixed structural correspondence, by preserving dynamics that remain meaningful across different target morphologies. To realize this, we propose a two-stage framework. Stage~I learns complementary multi-granularity abstract motion views and uses them to bootstrap cross-category video pairs that preserve transferable dynamics across diverse morphologies. Stage~II internalizes this supervision into direct reference-video-conditioned generation, removing the need for explicit motion extraction at inference. We further introduce OpenVMT-Dataset and OpenVMT-Bench for training and evaluating image- and text-conditioned motion transfer across Same, Near, and Far category gaps, and plan to release both upon acceptance. Extensive experiments demonstrate state-of-the-art motion fidelity and target preservation. Project page: https://miniz233.github.io/MotionBeyondMorphology/

Samantha Kelly / Bloomberg : Match reports Q2 revenue down 1% YoY to $853M, vs. $856.6M est., 13.3M paying users, vs. 13.4M est., forecasts Q3 revenue below est.; MTCH drops 8%+ after hours —  Match Group Inc. gave a revenue forecast for the current quarter that narrowly missed analysts' estimates …

Sam Sabin / Axios : The UK AISI says it observed a total of 19 instances where Mythos and GPT-5.6 Sol tried to hack people and companies during a routine cyber evaluation in July —  The U.K. AI Security Institute said it observed nearly 20 instances of Anthropic and OpenAI's most advanced models trying …

Maria Curi / Axios : Sources: the US' AI framework excludes open models and defines a covered frontier model as closed source with SOTA capabilities and national security risks —  The White House is excluding open models from its framework to test advanced AI capabilities, sources familiar with the matter told Axios.

Micah Maidenberg / Wall Street Journal : SpaceX says its Q2 capex was $18.4B, up from $2.8B a year ago, including $15.8B for AI —  SpaceX laid out $18.4 billion in capital investments during the April-to-June quarter, with most of the spending supporting artificial-intelligence efforts.  —  In its earnings report Tuesday …

Nathan Bomey / Axios : SpaceX reports Q2 revenue of $4.29B from its connectivity division, which includes Starlink, $100B of cash and marketable securities, and a $47.5B order backlog —  SpaceX exceeded revenue expectations and reduced its losses in its initial quarter as a publicly traded company …

Jonathan Vanian / CNBC : Pinterest reports Q2 revenue up 18% YoY to $1.18B, vs. $1.15B est., MAUs up 11% to 640M, forecasts Q3 revenue in line with estimates; PINS drops 8%+ after hours —  Pinterest shares fell 7% in extended trading on Tuesday after the company reported better-than-expected earnings and revenue but issued lukewarm sales guidance.

Blake Brittain / Reuters : A US appeals court overturns a ruling that had temporarily barred Perplexity from using its agentic shopping tools on Amazon's platform —  A U.S. appeals court on Tuesday overturned a ruling that had temporarily barred Perplexity from using its AI-powered agentic shopping tools on Amazon's (AMZN.O) platform.

Ian King / Bloomberg : AMD reports Q2 revenue up 50% YoY to $11.5B, vs. $11.3B est., Data Center revenue up 107%, forecasts Q3 revenue below some estimates; AMD drops 7%+ after hours —  Advanced Micro Devices Inc. disappointed investors with its latest sales outlook, a sign shareholders expected more of a return from the global expansion of AI data centers.

Jay Peters / The Verge : EA says its $55B acquisition by PIF, Silver Lake, and Affinity Partners has closed, taking the company private; the deal includes $20B in debt financing —  The massive deal has closed. The massive deal has closed. … Electronic Arts has officially become a private company.

Yoolim Lee / Bloomberg : Samsung previews zHBM, which vertically stacks HBM atop AI accelerators, and zNAND-O, built on V-NAND, and unveils the industry's first V10 BV-NAND architecture —  Samsung Electronics Co. offered a glimpse of its most advanced memory hardware, touting improved performance and power efficiency …

Bloomberg : SpaceX reports Q2 revenue up 92% YoY to $7.8B, vs. $6.81B est., and AI segment operating loss of $1.26B, vs. $2.39B est.; SPCX drops 5%+ after hours —  SpaceX's revenue exceeded Wall Street's expectations in the company's first quarterly financial report following its blockbuster initial public offering in June.

ABC News : Sources: possible cyberattacks targeting water and wastewater utilities have now been reported in at least 12 US states, and Iran is the prime suspect —  So far, there's been no widespread disruptions to water supplies or treatment.  —  Luke Barr, Jack Date, Katherine Faulders, Josh Margolin, and Aaron Katersky

Emma Roth / The Verge : Pavel Durov says an extortionist planted “illegal porn in a public chat” to get the app pulled from the App Store and manipulated Apple into “overreacting” —  Pavel Durov claims attackers post illegal content and manipulate reports to ‘trigger the removal of legitimate communities.’

Mistral AI Blog : Mistral releases Shieldstral, a 3B multimodal safety classifier that it says matches models up to 7x its size on text safety, available under Apache 2.0 —  Every product that ships a model needs to answer questions like these — but the right answer depends on the product, the audience, and the moment.

Maria Curi / Axios : Sources: the White House doesn't plan to publicly release its voluntary AI framework; details will be available to the companies that are part of the process —  The White House does not plan to publicly release its new framework for evaluating advanced AI models, three sources familiar with the discussions told Axios.

主要 PC 制造商惠普、华硕和宏碁开始少量使用长鑫的内存芯片。多家大型 PC 制造商已于今年年中完成了长鑫 DRAM 芯片的认证流程，开始在笔记本电脑中少量使用。由于长鑫优先向华为等国内客户供应内存芯片，因此其它厂商的供应量有限，相关笔记本电脑型号主要销往美国以外市场。PC 厂商对使用长鑫内存十分谨慎，因为他们担心会惹恼三大内存芯片制造商美光、三星和 SK海力士，这三大公司占据了逾九成的内存芯片市场。长鑫的内存并不比美光或三星等公司便宜，厂商也无法采购更多内存。

30 米望远镜（Thirty Meter Telescope，TMT）项目于 2014 年开始建造，计划 2027 年投入运行。望远镜选址定在夏威夷的 Mauna Kea 山，而 Mauna Kea 被当地原居民视为圣地，由于原居民的反对望远镜项目从 2015 年起处于停工状态，至今已超过 10 年。现在西班牙正试图在该国的加那利群岛建造 30 米望远镜，它提出了 11.4 亿美元的方案用于建造和未来的运营费用。

开源多媒体库 FFmpeg 9.0 "Lei" 释出。新特性包括：Vulkan APV 视频解码和 Apple ProRes RAW Vulkan 加速、Vulkan v360 视频滤镜、HE-AAC 960 解码、NVIDIA CUDA 转置滤镜、动画 WebP 解码和解复用（demuxing）、AMD AMF 增强、AVX-512 优化等。其它包括 扩展 AMF 色彩转换器 (vf_vpp_amf) 的 HDR 功能、MP4 复用器支持 LCEVC 音轨复用，等等。

一项研究发现，11-12 岁设立第一个社媒账户的学生，其数学和阅读测试成绩低于几年后才设立账户的学生。16 岁时两组学生的成绩差距相当于六个月的学习差距。研究人员分析了逾 5000 名意大利中小学生的社媒使用习惯调查，及其在标准化数学、意大利语和英语考试中的成绩。13-14 岁时接受测试的学生中在 11 或 12 岁就设立社媒账户的学生，成绩低于至少等到 14 岁才开始使用社媒的学生。15-16 岁时意大利语的成绩差距依然存在，数学成绩差距则进一步扩大。英语成绩没有受到影响，研究人员认为是社媒上的英语内容较多，为学生们提供了偶尔学习的机会。研究人员认为，频繁查看社媒对学生而言是一大干扰因素。

云南玉溪的研究人员在玉溪红塔区的一个洞穴内发现了一种新的蜗牛，以《山海经》中的山神名字命名为帝江或 Erhaia dijiang。Erhaia dijiang 完美适应了终年无光的洞穴环境，是一种典型的洞穴专生物。它在长期演化过程中完全失去了眼睛，眼部结构已彻底退化。其贝壳小巧呈圆锥形，高度透明，身体柔软且没有色素，触角细长。实验室观察发现，该物种极度畏光，一旦受到光照便会迅速钻入岩石缝隙中躲避。Erhaia dijiang 生活在海拔2062米的洞穴中，刷新了洱海螺科所有已知洞穴物种的海拔分布纪录，证明洞穴淡水螺能够在海拔超过 2000 米的喀斯特洞穴中生存。该物种栖息于水深不足 1 米的洞穴水体底部砾石和碎石之间，洞内水温常年稳定在约 20℃。雨季水流湍急时，偶尔会有少量个体被冲出洞口，但它们从未在附近地表溪流中被发现，活动范围始终局限于距洞口约两米以内。在实验室条件下，它们主要取食岩石表面的硅藻，也能够摄食人工投喂的鱼饲料。

越来越多的动物研究表明，限制蛋白质摄入量或许能够延缓衰老、延长寿命。那么，这是否意味着摄入过多蛋白质存在危害呢？威斯康星大学麦迪逊分校的 Dudley Lamming 表示：“我们通常认为膳食蛋白质是有益的。”但他指出，最新的证据显示，“对于久坐不动的人或动物来说，如果摄入的膳食蛋白质超过所需的量，会引发代谢问题，还可能缩短寿命”。也有学者指出，在小鼠、大鼠等动物身上开展的营养学研究并不适用于人类，而且高蛋白摄入量与人体的益处相关。Lamming 认为，综合 300 项研究，当下流行的增加蛋白质摄入的做法，对经常运动的人而言或许并无不妥，因为他们需要更多蛋白质来修复肌肉损伤，然而摄入过多蛋白质可能对久坐不动的人造成伤害。

戴森球一直都被列为地外文明技术特征的典型标志，天文学家使用韦伯望远镜的中红外仪器（MIRI），对两颗因红外超出而被列为戴森球头号候选的 M 型矮星进行了复查，结果发现所谓异常根本不在恒星身上，而是来自天空中与恒星相距仅约 1 角秒的两颗毫无关联的遥远星系。这意味着两个最有希望的“外星文明巨型工程”均为天然天体物理源。戴森球概念由物理学家弗里曼·戴森在 1960 年提出，其认为一个足够先进的文明为突破行星能源瓶颈，会转向母恒星收集能量，而任何能量收集结构都无法隐藏废热，必然在中红外波段留下特征性辐射。正因如此，NASA 将戴森球列为地外文明典型技术特征之一，并通过望远镜寻找这一先进文明巨型工程的可能痕迹。基于该假设，能在数百万颗恒星中筛出红外超出目标，即可把它当作可能存在外星文明的线索。

在 LG 之后，三星也宣布禁止嵌入了住宅代理得智能电视应用。嵌入住宅代理 SDK 的应用会出售用户的家用 IP 作为代理服务使用。智能电视是理想的代理主机，它基本上一直处于插电状态，同时接入了家用 WIFI，但不像 PC 没人会去检查其可疑后台活动。此前的研究发现，LG webOS 应用商店逾 42% 的游戏和其它应用内置了住宅代理 SDK，三星 Tizen 应用商店也有逾四分之一应用内置了住宅代理 SDK。部分嵌入住宅代理 SDK 的应用安装量多达数亿，相关应用甚至还被三星推荐过。三星在一份声明中表示正在封禁嵌入住宅代理 SDK 的应用，将移除包含此类功能的应用。

韩国录得自 1904 年现代气象观测以来的最高气温，韩国东南部的庆尚南道梁山市周日下午气温达到了 42.5℃，连续第五天超过 40℃。当局警告居民“立即停止所有户外活动”。自 1973 年以来，韩国白天气温连续 3 天以上达到 40℃ 的情况仅出现过三次。韩国逾 20 个地区周日发布了高温紧急预警——这是今年新增的预警级别，旨在更好应对不断上升的气温。当预计高温地区体感温度将达到 38℃或实际温度达到 39℃ 时，会发布高温紧急预警。气象部门敦促民众“立即停止所有户外活动”，称“没有空调的室内空间很危险”。气象部门建议：“立即前往阴凉处，如指定的避暑中心或阴凉区域，在休息时保持水分充足。”韩国气象厅的数据显示，过去五年韩国每年平均高温天气天数增加了一倍多达到 19 天。

苹果在恰当的时机推出了一款价格合适的产品，由于内存因供应问题而价格暴涨，美国学校以前采购的教育笔记本电脑 Chromebook 价格也随之大幅增长，苹果推出的 500 美元教育笔记本电脑 MacBook Neo 变得价廉物美。苹果最新财报电话会议披露，美国教育机构正大量采购 MacBook Neo，近半是为了替代 Chromebook 等竞品。苹果举例说，佛罗里达 Pinellas County 学区 18 所高中 2.5 万名学生正从 Windows 迁移到 MacBook Neo；华盛顿州 Peninsula 学区 401 逾 8000 名学生从 Chromebook 迁移到 MacBook Neo；俄克拉荷马州 Midwest City-Del City 学区采购了逾 6000 台 MacBook Neo。

微软 Windows 和设备执行副总裁 Pavan Davuluri 通过官方博客谈论了该公司在 Windows 11 上持续展开的改进工作，虽然大部分工作尚未推送给用户。他透露接下来的一项工作是优化 Windows 11 在 8GB 内存以上配置下的运行效率。由于目前的内存供应危机，PC 厂商又开始提供 8GB 内存配置的电脑，然而 Windows 11 在 8GB 内存下只能勉强运行，运行其它程序面临很大的局限性。Davuluri 表示该公司将减少 Windows 的内存占用，在用户日常使用的 PC 上提供快速响应的 Windows 体验。

塑料颗粒显然已经遍及地球的各个角落。根据发表在《Water Research》期刊上的一项研究，韩国研究人员检查了生活在水下 2000 米处深海热泉附近的蜗牛和贻贝，12 只受检动物 92%（11 只）体内都发现了微小的塑料碎片。海洋生物学家 Se-Joo Kim 说，深海热泉曾被认为是地球上最与世隔绝的环境之一。每只深海热泉动物体内平均发现了 3.42 个塑料碎片，最常见的是聚苯乙烯。研究人员还发现，摄食行为会影响微塑料在动物体内的积累位置。海床进食的蜗牛，塑料聚集在消化器官中；滤食性贻贝的塑料分布则更为均匀。

澳大利亚联邦银行、微软、 Uber 和凯悦酒店等公司正用自动化聊天和电话系统去处理以前需要人工完成的工作。AI 系统开始大幅削减客服人数。呼叫中心行业在美国、印度和菲律宾等地雇佣了数百万从业者。随着 AI 技术的进步以及高管面临拥抱新技术的压力，企业日益用生成式 AI 工具取代客服。美国的客服务人数正在下降，未来这一趋势会继续下去。分析师估计到 2030 年近半客服岗位将受到影响。西方公司此前将客服外包到熟悉英语的菲律宾、印度等国，随着 AI 的普及，这些国家的客服人数预计会大量减少。澳大利亚最大银行——联邦银行（Commonwealth Bank of Australia）裁减了数百名客服，每年预计节省数千万美元。微软将其客服团队（其中包括合同工和全职员工）从约 5 万人缩减至 4 万人。微软销售和服务运营负责人 Judson Althoff 在 4 月曾表示，AI 每年为公司节省约 7.5 亿美元的客服成本。他表示，更复杂的问题仍需要人工支持，但公司正不断扩展自动化修复的范围。凯悦酒店去年裁掉了美洲地区三成的内部客服。Uber 裁减了 10% 的客服岗位以“拥抱 AI”。

澳大利亚政府为 16 岁以下儿童的社交媒体禁令辩护，称禁令重塑了全国的舆论，极大改变了与家长之间的对话，关闭了数百万​​个儿童账户。政府从未指望禁令能得到 100% 的遵守，就像最低饮酒年龄法无法做到 100% 遵守一样，但法律的存在仍然是恰当的。此前的研究发现，禁令生效三个月后，逾八成澳大利亚青少年仍在继续使用社交媒体，10-15 岁儿童在 3 月使用社交媒体的频率与 12 月 10 日禁令生效前一样高。

在 Linux 游戏社区多年请求之后，GOG Galaxy 游戏启动器将原生支持 Linux。GOG 有游戏的 Linux 版本可以下载，但只能通过网站下载，GOG Galaxy 客户端只支持 Windows 和 macOS 操作系统。GOG 没有公布支持 Linux 的时间，只是表示正在开发中。GOG Galaxy 类似 Steam 客户端，支持自动游戏安装和更新、云端存档同步、成就追踪、游戏时长统计、游戏库管理、商店浏览、 好友列表和社交功能等等。在 Valve 的 Proton 推动下，Linux 对游戏的支持日益改进，越来越多的游戏公司也认识到了 Linux 的重要性。