OrangeBot.AI Digest — 2026-07-19

Fell for a discount again?

The open source Ahrefs alternative

Every place worth the detour, on one interactive map

See it once. Ask forever.

6.4x faster than llama.cpp, 3.9x faster than MLX

Privacy-friendly web analytics with traffic and revenue.

Turn your SaaS into a clickable demo in 90 seconds.

An AI coworker with its own cloud computer

Build landing pages with Aceternity UI

Learn Japanese, one prefecture at a time.

Turning fragmented knowledge into one simple system

A markdown editor you and your agent co-edit

An experimental synth for playable generative sound

Open-source alternative to Storylane, Navattic, and Arcade

The data layer for AI agents

A growing gap separates inference context lengths from RL post-training: inference systems are approaching million-token contexts, while post-training workloads often remain at 256K tokens or below and rely on length generalization at deployment. The gap is especially important for AI agents, whose observations, tool outputs, documents, and prior decisions accumulate over long trajectories. LongStraw is an architecture-aware execution stack for million-token RL post-training under a fixed GPU budget, instantiated with Group Relative Policy Optimization (GRPO). It evaluates the shared prompt without autograd, retains only model-specific state needed by later tokens, and replays short response branches one at a time, reducing the live training graph at the cost of additional replay time. We implement it for the hybrid recurrent and full-attention Qwen3.6-27B and the compressed-attention mixture-of-experts GLM-5.2. On eight H20 GPUs, LongStraw completes grouped Qwen scoring and response backward at 2.1M positions for groups of 2 and 8; increasing the group size adds only 0.21 GB of peak allocated memory, while a separate stress test reaches 4.46M positions. On 32 H20 GPUs, we validate the end-to-end LongStraw execution path for a 2.1M-token prompt across all 78 layers of GLM-5.2. These experiments establish execution capacity rather than complete training correctness because the captured prompt state is detached and some distributed forward and gradient composition paths remain incomplete.

Recent advances in video understanding have spanned motion, long video, and streaming interaction, driving this field toward real-world applications. Despite this progress, current open-source models remain limited in several ways. They often struggle to generalize across diverse video types, making them effective only in specific domains. High computational demands further restrict their efficiency and scalability. Moreover, most models are only partially open, with key components such as training code, strategy, or datasets unavailable, which hinders reproducibility and slows community-driven development. To address these issues, we introduce VideoChat3, a fully open, efficient, and generalist video-centric MLLM. VideoChat3 advances video understanding through two complementary designs. For efficiency, we introduce Inflated 3D Vision Transformer (I3D-ViT) and Adaptive Frame Resolution for Streaming Video Perception, which enables efficient spatiotemporal representation and reduces the cost of processing video inputs during training and inference. For effectiveness, we develop a scalable video data synthesis pipeline that curates three diverse, high-quality training datasets: VideoChat3-Academic2M, VideoChat3-LV116K, and VideoChat3-OL617K, covering general, long-form, and streaming video scenarios, improving the model's generalization across domains. By integrating these designs, VideoChat3 achieves a rare balance of broad generalization and computational efficiency. Experiments across general, long-form, and streaming benchmarks demonstrate that VideoChat3 surpasses prior open-source models with equal or larger parameter counts with only 4B parameters and higher efficiency.

Large language models are increasingly trained as interactive agents for long-horizon tasks involving multi-turn interaction, tool use, and environment feedback. Outcome-based reinforcement learning (RL) provides a practical optimization paradigm, but its sparse trajectory-level rewards offer limited guidance on intermediate decisions, leaving a supervision gap between episode-level outcomes and token-level policy learning. We propose SEED (SElf-Evolving On-Policy Distillation), a self-evolving framework that converts completed on-policy trajectories into training-time hindsight skills and distills their behavioral effect back into the policy model. SEED first fine-tunes the policy to analyze completed trajectories and generate natural-language skills that capture reusable workflows, decisive observations, or failure-avoidance rules. During RL, the current policy both collects trajectories and serves as the analyzer that extracts hindsight skills from them. Policy updates therefore improve subsequent decision making and skill analysis together, allowing hindsight supervision to evolve with the policy. SEED then re-scores the sampled actions under ordinary and skill-augmented contexts, converting the skill-induced probability shift into a dense token-level on-policy distillation signal. This signal is jointly optimized with outcome-based RL, keeping the auxiliary supervision aligned with the current trajectory distribution. Extensive experiments on text-based and vision-based agentic tasks show that SEED consistently improves performance and sample efficiency, exhibiting robust generalization to unseen scenarios. Our code is available at https://github.com/jinyangwu/SEED.

Recent advances in Tool-Integrated Large Language Models have made web search a core capability of information-seeking agents. However, as interaction histories grow, agents increasingly struggle to track task progress. When search attempts fail to yield useful evidence, current single- and multi-agent systems can become trapped in repetitive loops, wasting search budgets and ultimately compromising the quality and completeness of the final output. We introduce SearchOS, a system-level multi-agent framework that turns fragile, implicit search progress into explicit, persistent, and shared state. First, we formulate open-domain information seeking as relational schema completion with grounded citations, where agents discover entities, populate attributes across linked tables, and anchor each value to source evidence. Then we design Search-Oriented Context Management (SOCM), which externalizes the evolving state into Frontier Task, an Evidence Graph, a Coverage Map, and Failure Memory. Built on SOCM, SearchOS applies a pipeline-parallel scheduling mechanism that overlaps the execution of sub-agents and continuously refills freed slots with tasks targeting unresolved coverage gaps to improve utilization and throughput. To schedule and control the execution of search agents, SearchOS introduces a Search Tool Middleware Harness that intercepts model and tool interactions to record grounded evidence and react to stalls or budget exhaustion, and provides a reusable hierarchical skill system comprising strategy and access skills to augment the agents' search process and avoid repeating failed search patterns across runs. On WideSearch and GISA, SearchOS leads all metrics among the evaluated single- and multi-agent baselines, paving the way toward robust information-seeking collaboration.

World-action models (WAMs) are emerging as a promising foundation for embodied control: rather than predicting actions alone, they learn representations that couple action generation with future world prediction. This coupling is often viewed as a source of robustness, interpretability, and safety, as a robot's action can in principle be checked against its imagined future. In this paper, we show that this assumption is fragile. We introduce BadWAM, a unified framework for modeling and evaluating World-Action Drift Attacks: a new class of WAM-specific adversarial attacks that use small visual perturbations to break the alignment between what a WAM imagines and what it executes. BadWAM characterizes this attack surface along two natural criteria: attack strength and stealthiness. When the adversary prioritizes disruption, BadWAM instantiates an action-only adversarial attack, which directly drives the model toward task-failing actions. When the adversary additionally prioritizes stealth, BadWAM instantiates an imagination-preserving adversarial attack, which seeks to induce harmful action shifts while keeping the model's predicted future close to its clean imagination. Together, these two attacks capture a spectrum of WAM-specific failures: from overt action hijacking to stealthier cases where the model appears to imagine a plausible future but executes a desynchronized action. We evaluate BadWAM across different variants of WAMs. Results show that our attacks substantially reduce task success rates under closed-loop execution. For example, our action-only attack reduces the model performance from 96.5% to 43.1% success. The results of our imagination-preserving attack further exposes a WAM-specific vulnerability: moderate future-preserving regularization can maintain strong attack performance while reducing future imagination drift.

Video generation increasingly relies on keyframe-based workflows, where creators specify a sequence of reference images to guide generation. Although recent models support multi-keyframe conditioning, it remains unclear whether they can faithfully reproduce the prescribed keyframes while maintaining overall video quality. We present KeyFrame-Compass, the first comprehensive benchmark for evaluating keyframe-conditioned video generation. The benchmark contains 386 carefully curated samples spanning three application domains, two video structures, two prompt granularities, two conditioning formats, and four keyframe densities, enabling controlled analysis under diverse generation settings. We further introduce an automated evaluation framework that jointly measures keyframe execution and overall video quality. Specifically, we decompose keyframe execution into six complementary metrics covering presence, fidelity, temporal ordering, localization, persistence, and uniqueness, while assessing overall video quality through evidence-grounded MLLM judgments augmented with specialized perception models. Experiments on nine representative video generation systems reveal several fundamental limitations. Current models exhibit a clear trade-off between faithful keyframe execution and natural video synthesis. Their performance further degrades as keyframe constraints become denser and most open-source models also fail to interpret storyboard-grid inputs as temporally ordered keyframe sequences.

Multi-reference-to-audio-video (MR2AV) generation aims to generate coherent audio-video content conditioned on multiple references and textual instructions. Existing benchmarks mainly focus on text-driven generation, single-reference subject preservation, or isolated audio-video alignment, leaving the emerging MR2AV setting largely unexplored. Compared with these settings, MR2AV requires models to jointly reason over multiple references while generating synchronized visual and audio content. Models must not only preserve each reference faithfully but also correctly bind and compose multiple referenced entities into coherent audio-visual events. To address this gap, we introduce MultiRef-Compass, a unified benchmark for MR2AV generation. It comprises 350 carefully curated samples constructed through a scalable and controllable asset-composition pipeline, covering multi-view subject preservation, multi-entity binding, and human-object-scene composition. To provide interpretable assessment, MultiRef-Compass defines an evaluation protocol with four dimensions: Basic Quality, Reference Consistency, Audio-Visual Consistency, and Instruction Following, using 14 sub-metrics. MultiRef-Compass integrates automatic metrics with a rejudging-enhanced MLLM-as-a-Judge framework, enabling scalable and auditable evaluation of both perceptual fidelity and reference-conditioned composition. Extensive experiments on eight representative MR2AV systems reveal substantial room for improvement across multiple evaluation dimensions, underscoring the need for a comprehensive benchmark and positioning MultiRef-Compass as a foundation for future MR2AV research.

Building interactive worlds that respond coherently to player actions has long been a shared goal of computer graphics, games, and artificial intelligence. Recent video generative models provide a data-driven route toward this goal by predicting future observations conditioned on user actions, and are increasingly regarded as potential next-generation game engines. Realizing a genuinely interactive game world, however, requires interaction outcomes that follow rules over evolving game conditions, consequences that persist over long horizons, and a generation loop that operates in real time. Conventional game engines realize these properties through a recurrent action-state-observation loop, in which player actions update an explicit game state according to predefined rules and observations are rendered from the resulting state. Taking this loop as an organizing lens, this paper examines interactive game world modeling along four dimensions: player action control, game state dynamics, state-observation persistence, and real-time interactive generation. For each dimension, we start from the capabilities required by an interactive game world, group existing approaches into representative families, and discuss the strengths and trade-offs of each family. Complementing this analysis, we present a scalable data engine for Black Myth: Wukong that collects over 90 hours of gameplay with frame-aligned player actions, ground-truth game states, and visual observations, together with structured and semantic annotations, as a resource for state-aware game world modeling. We hope this paper offers a clear picture of where the field stands and fosters progress toward interactive game worlds.

Human cognition does not separate understanding and generation. A teacher at a whiteboard speaks and draws together, each modality reshapes the other. In this paper, we bring this coupled loop to artificial systems. Masked Diffusion Models (MDMs) are ideally suited to this task, yet existing samplers either decode text and image interleavedly or independently update them in parallel branches that share only previous-step history, but not the other modality's latest decisions within the same step; combined with MDMs' inability to remask, cross-modal contradictions are neither detected nor repaired. We introduce Self-Correcting Coupled Markov Jump Processes (SC-CMJP), a framework in which one modality's transition rates are functionals of the other modality's confidence score, as weighted by cross-modal attention. Furthermore, a remasking jump retracts commitments the moment cross-modal evidence turns against them. In conjunction with SC-CMJP, we introduce CO_2Jump (Self-text{CO}rrecting text{CO}upled text{Jump}), a novel training-free single-pass sampler for joint multimodal geneneration. For training and evaluation purposes, we have created and will release three large-scale joint multimodal generation corpora: JEdit-1M, JMaze-200K, JNono-200K, with matching in- and out-of-distribution benchmarks. CO_2Jump achieves best joint performance for image understanding and editing as well as visual reasoning (maze and nonogram solving). The performance of the sampler scales monotonically with the number of denoising steps, evidence that the benefits of cross-modal coupling compound across the trajectory. Project page: https://coupled-jump.github.io

Learning broad world knowledge directly from raw visual data is a fundamental capability of intelligence. We introduce UniVR, the first investigation into simultaneously learning complex reasoning, fine-grained physical dynamics, and long-term planning from pure visual demonstrations. At its core, UniVR features VR-GRPO, a reinforcement learning paradigm with complementary global and step-level rewards. This approach enforces logical coherence and physical consistency throughout the reasoning process without requiring task-specific heuristics or image-text pairs. To train and evaluate UniVR, we construct VR-X, a large-scale benchmark curated from 16 diverse sources spanning long-horizon manipulation, spatial puzzles, and physical reasoning. It is the first comprehensive suite to assess these heterogeneous capabilities under a purely visual protocol. Remarkably, UniVR achieves up to a 25% improvement on VR-X, and its superior visual reasoning also boosts performance on various multimodal understanding benchmarks. These findings underscore the vast potential of reasoning within visual spaces, with all code, data, and models are open-sourced for further research.

Embodied cognition requires agents to connect high-level task reasoning with the physical states to be achieved. We introduce Hy-Embodied-RxBrain, an embodied cognition foundation model with joint language-visual reasoning and imagination. Unlike vision-language models that emphasize scene understanding and textual decision making, or generative world models that mainly predict future visual states, RxBrain represents embodied plans in a single planning sequence where language and visual imagination play complementary roles. Language provides the abstract structure of a plan, including task decomposition, planning primitives, constraints, temporal order, and decision logic, while visual imagination grounds this structure through world state prediction and joint subgoal planning, associating each planning step with intermediate and final physical states. RxBrain adopts a unified multimodal Mixture-of-Transformers architecture that supports language, image, and video understanding and generation within one model. To train this capability, we build an automatic pipeline that converts embodied videos into joint text-visual planning supervision by decomposing videos into planning steps and aligning them with visual state transitions. We further introduce RxBrain-Bench to evaluate whether models can represent embodied plans through joint textual and visual components rather than separate understanding or generation. Experiments show that RxBrain maintains embodied understanding and generation abilities, and produces plans with coupled textual reasoning, world state prediction, and joint subgoal planning. We also extend RxBrain to continuous robot action generation, where it shows promising real-robot performance without large-scale action-data pretraining. These results provide an initial step toward foundation models for embodied cognition.

Reinforcement learning has become a standard post-training recipe for large language models, but dense full-parameter updates create two deployment-relevant bottlenecks: suppressed reasoning performance, often reflected by premature saturation of test-time scaling, and interference when consolidating multiple capabilities through multi-domain training or model merging. We show that the reasoning-effective component of these updates is largely concentrated in the base model's spectral space, motivating Subspace-Aligned Rewiring (SAR), a post-hoc editing method that retains this spectral core while removing orthogonal components. SAR therefore preserves reasoning gains and filters residual update directions that suppress performance or amplify cross-domain interference. Across several model families and scales, SAR extracts compact reasoning cores using as little as approximately 0.58% of total parameters: it preserves over 99% of post-training performance and improves high-k exploration in mathematical reasoning, and generalizes to agentic coding by improving six of seven open benchmarks on an in-house model. SAR also purifies mixed-domain training updates by releasing suppressed coding capability while maintaining math reasoning and instruction following. It further enables model merging across experts, yielding cross-domain generalization that surpasses previous merging baselines and even the best single-domain experts. Overall, SAR shows that extracting reasoning-effective updates from parameter geometry can serve as a training-free mechanism to improve reasoning and multi-domain performance.

On-policy distillation (OPD) has become a key paradigm in LLM post-training, yet its training dynamics remain poorly understood. We present a systematic study examining the role, pathologies, and regulations of OPD. We first clarify the role of OPD as an exploration catalyst: it steers the student toward correct reasoning paths via dense token-level guidance, without expanding capability ceiling. We confirm this by showing that prompt diversity matters more than per-problem sampling numbers, and critically, that the effectiveness of OPD hinges entirely on the quality of its guiding signal. This dependency exposes two pathologies that derail exploration. The Student-Teacher Mismatch occurs when a large teacher-student distributional gap causes the guiding signal to misalign with task correctness, steering exploration in counterproductive directions. Length Exploitation arises when the aggregated token-level objective creates length-dependent shortcuts, allowing the student to game the reward landscape through response truncation or redundant padding, exploring degenerate length modes rather than reasoning strategies. To tame these pathologies, we investigate lightweight signal regulations: advantage clipping and log-scale compression, ensuring exploration is guided by faithful signals. Experiments across seven benchmarks demonstrate that these regulations alleviate length exploitation and enable effective distillation, stably surpassing OPD variants and RLVR baselines, thereby confirming that well-regulated signal quality, rather than mere teacher scale, governs successful exploration in OPD.

Recent robot foundation models operate with single-step or short-history visuomotor context. We introduce Test-Time-Training Robot Policies (RoboTTT), a robot model and training recipe that scale visuomotor context to 8K timesteps, three orders of magnitude beyond state-of-the-art policies, without growing inference latency. At this context length, we unlock new robot capabilities: one-shot in-context imitation from human video demonstrations, on-the-fly policy improvement, robustness to perturbations, and stronger performance on multi-stage, long-horizon tasks. We also observe, for the first time, steady gains in closed-loop performance as pretraining context length scales. At its core, RoboTTT integrates Test-Time Training into robot foundation models such as Vision-Language-Action policies, yielding a sequence model whose recurrent state consists of fast weights, parameters updated by gradient descent during both training and inference, compressing histories into weight space and retrieving contextual information for long-context conditioning. To scale training context length, the recipe combines sequence action forcing with truncated backpropagation through time. On challenging real-robot manipulation tasks, RoboTTT improves overall performance by 87% over the single-step context baseline and fully completes a five-minute, ten-stage assembly task, which no baseline ever does. RoboTTT trained with 8K-timestep context outperforms the same model pretrained with 1K timesteps by 62%, suggesting context length as a new scaling axis for robot foundation models. Videos are available at https://research.nvidia.com/labs/gear/robottt/

We present Wan-Streamer v0.3, which reframes our native-streaming interaction model under a single organizing view: a video is a world plus an event stream. The world is the persistent context in which a video unfolds, including the environment, scene, subjects, ambient acoustic conditions, voice characteristics, and other relatively stable conditions. The event stream is everything that changes over time within that world, including scene or environmental changes, subject behavior, speech, and other sounds. This yields a general-purpose pretraining task over large amounts of real video: given a world and incoming input, predict how the world moves, changes, and responds in real time. The resulting competence can be specialized to a broad family of real-time downstream tasks. We instantiate it on real-time full-duplex audio-visual interaction, where the event stream is the agent's speech together with free-form behavior. Functionally, the model's multimodal understanding process is vision-language-action-like: it maps multimodal user input to language-form speech and behavior actions. Wan-Streamer v0.3 preserves the v0.2 operating point: 640x368 video at 25 FPS, a 160 ms streaming unit, approximately 200 ms model-side response latency, and approximately 550 ms total interaction latency under a 350 ms bidirectional network budget.

@kimi_moonshot : Moonshot AI pauses new subscriptions, saying that over the past two days Kimi K3 demand nearly passed the limit of its capacity, and also reworks its plan tiers —  Kimi K3 has received far more love than we expected, and our GPUs are feeling it. Over the past 48 hours, demand has pushed close to the limits of our current capacity. To protect the experience of existing subscribers, we're temporarily pausing new subscriptions and

Belle Lin / Wall Street Journal : As many companies cut AI costs by using cheaper models, some, like Shopify, go all-in on frontier models, barring engineers from using anything else —  Despite steep and rising price tags, some companies heavily favor the more powerful ‘frontier’ AI systems over cheaper alternatives

Lauren McCarthy / New York Times : Kalshi and Polymarket bets on the FIFA World Cup final top $5.69B; Dune Analytics says total wagers on the platforms surpassed $50B for the first time in June —  When Argentina and Spain take the field on Sunday for the World Cup final, more than $5.69 billion will be riding on the winner …

Mark Gurman / Bloomberg : Sources: Apple is piloting a system in Apple Stores called Live Notes that lets Genius Bar staff make transcripts and summaries of conversations with customers —  Also: The company tries out AI recording for Genius Bar appointments.  —  This week, we'll look at why Apple's massive lawsuit …

Bloomberg : Alibaba launches a 2.4T parameter Qwen3.8 Max preview that it says rivals frontier AI models and is second only to Fable 5, plans to make it “open-weight soon” —  Alibaba Group Holding Ltd. launched a preview version of its flagship Qwen3.8 Max model, which it described as comparable …

Andrew Hill / Financial Times : AI is reshaping entry-level professional services jobs, as companies redesign hiring, training, and workplace culture rather than simply cut junior roles —  Here are the ways leading companies are already responding to the AI revolution in professional services

Bloomberg : Sources: Kimi developer Moonshot told investors it is preparing for a Hong Kong IPO in as early as six months; its ARR hit $300M in June, up from $200M in April —  Moonshot AI has told investors it's preparing to list in as early as six months, seizing the opportunity to tap capital markets …

Alexandra Lindsay / The San Francisco Standard : Analysis: OpenAI and Anthropic employees are donating to campaigns more heavily and cohesively than Google, Meta, and Airbnb employees did post-IPO —  San Francisco is still waiting for the AI IPOs — the wave of money from Anthropic and OpenAI millionaires and billionaires that is likely …

Zusha Elinson / Wall Street Journal : A profile of businessman Sebastian Rucci, whose past ventures were subject to legal probes, as he bids to build California's biggest data center, a $10B project —  Sebastian Rucci's past has become a theme in opposition to his proposal.  He counters: 'Do I look like a guy that's going away?'

Haley Zimmerman / Financial Times : Big US pizza delivery chains are struggling as apps such as DoorDash and Uber Eats give independent pizzerias and other restaurants greater market access —  Large corporate chains lose their edge as services such as DoorDash and Uber Eats level the playing field

Wall Street Journal : Sources: CIA operative Jonny Gannon spied on G42 to probe its China ties and helped the UAE gain expanded access to US AI chips by allaying US suspicions —  While conducting espionage against powerful sheikh's tech company, Jonny Gannon helped Gulf country quash Washington suspicions about China ties

Ann Cao / South China Morning Post : Alibaba open-sources its chip software, following similar plays from Huawei and Moore Threads, as Chinese GPU makers try to break the dominance of Nvidia's CUDA —  The firm's chip unit T-Head aims to lower migration barriers to Zhenwu AI computing architectures, following similar initiatives by Huawei and Moore Threads

9to5Mac : Analysis finds 60+ “jacket apps” on the App Store disguised as simple games and utilities that become gambling apps when accessed from Brazilian IP addresses —  An investigation by 9to5Mac reveals dozens of apps that disguise gambling platforms as simple games and utilities.

Joshua Cohen / Ars Technica : A look at AI's potential impact on insurance-coverage decisions like prior authorization as the Trump admin starts to pilot using AI to evaluate Medicare claims —  If you're like me, you or a loved one has struggled through the process of gaining pre-approval for the medical care that your physician has recommended.

Geraldine McKelvie / The Guardian : UK's incoming PM Andy Burnham is expected to scrap Keir Starmer's plans for digital ID cards; last year a petition opposing the cards attracted ~3M signatures —  Incoming PM will reallocate unspecified resources from unpopular plan to helping with cost of living

2024 年 3月 29 日，一位微软工程师在家执行例行测试时，发现登录测试机的时间比往常慢了约半秒。大多数人都不会认为这是什么大问题。但他却深入展开了调查，在流行压缩工具 XZ Utils 中发现了一个故意植入的隐蔽后门。有人花了两年时间才把后门植入到该工具中。Adrian Mastronardi 发表了一本关于 XZ 后门事件的书《Half a Second》，在非商业使用、禁止衍生的 CC 许可证（CC BY-NC-ND 4.0）下免费提供。本书讲述了 一位精疲力竭的志愿者独自维护着代码，被耐心而巧妙的操纵，最终交出了代码的维护权；一位工程师凭借一连串的运气和来之不易的直觉，在半秒钟的好奇心驱使下发现了这次攻击；以及构建该后门的幕后操纵者，此人至今身份不明，其身份可能永远也无法被揭露。

反数据中心抗议者在美国 42 州举行了 142 场抗议活动。这一抗议行动获得了两党民众支持、由前茶党领袖联合创办的组织 HumansFirst 协调。抗议者集会反对 HumansFirst 所称的数据中心“缺乏问责”的扩张以及“对自由不可接受的侵犯”。在弗吉尼亚州 Fredericksburg，一位年轻人举的牌子上面写着：“我们靠水而不是数据生存。”亚特兰大抗议者的标语是“佐治亚州从未投票支持超大规模人工智能数据中心”。HumansFirst 表示暂无全美和各大城市抗议者人数的统计数据。反对数据中心建设是少数几个能跨越意识形态团结美国人的议题之一。民调显示，只有三分之一的美国人认可数据中心的建设速度，只有 14% 的受访者支持在其社区建设数据中心。

法国持牌博彩/赌博游戏监管机构 Autorité Nationale Des Jeux 命令 ISP 屏蔽最大预测市场平台 Polymarket，任何被发现宣传该非法赌博网站的人都可能被处以最高 10 万欧元的罚款。ANJ 此前已经限制法国居民在 Polymarket 上进行任何金融交易，但它发现法国居民绕过了限制，Polymarket 的法国访客仍然持续增长。ANJ 希望加大对 Polymarket 的打击力度，再次强调该平台被认定为非法赌博网站。西班牙政府也下令封锁了预测市场平台 Polymarket 和 Kalshi。

2025 年美国新车仅有 0.6% 是手动挡，汽车已是自动挡的天下，电动汽车和自动驾驶汽车的兴起也加速了手动挡汽车的消失。相比下欧洲的情况略好，行业分析公司 JATO Dynamics 的数据显示，欧洲最大汽车市场手动挡汽车的注册比例从 2001 年的 91% 下降到 2024 年的 29%。曾以手动挡汽车知名的斯巴鲁（Subaru）已停止提供手动挡车型，大众汽车将于今年停产在美国销售的最后一款手动挡车型捷达 GLI。丰田、本田和宝马都减少了面向美国市场的手动挡车型。

天文学家确认一颗名为 GJ3378b 的系外行星，位于距离地球仅 25 光年的红矮星宜居带内，可能具备孕育生命的条件。这颗行星位于鹿豹座方向，围绕一颗暗淡的红矮星运行，被视为目前已知距离地球最近的潜在宜居世界之一。GJ3378b 最初于 2024 年被法国天文学家发现。当时测定的质量为地球的 5.26 倍，被认为是一颗类似微型海王星的气态行星。天文学家利用基特峰国家天文台的 WIYN 3.5米望远镜等设备进行二次观测后，修正了这一结论。新数据显示，GJ3378b 的真实质量仅为地球的 2.3 倍，属于岩石质地的“超级地球”。轨道参数同样得到修正。最初认为该行星的公转周期为 25 天，实际仅为 21 天。这意味着它比此前判断的更靠近母星，但也恰好处于宜居带内——这一区域允许行星表面温度维持液态水的存在。这颗行星从母星接收的辐射量约为地球从太阳获取的 90%，位置非常理想。

“电视会腐蚀你的大脑”这句话也许在字面意义上是正确的。根据发表在《Alzheimer’s and Dementia: Journal of the Alzheimer’s Association》上的一项研究，常看电视的人大脑中与记忆相关的区域体积缩小，额叶和枕叶变小，大脑白质出现了与衰老、中风风险、认知能力下降和痴呆症相关的损伤区域。研究人员分析了约 1700 名成年人的数据，他们的平均年龄 53 岁，在 1987-1989 年间参与了心血管研究 Atherosclerosis Risk in Communities(ARIC) 。参与者被询问了休闲时看电视的频率。20 多年后参与者接受了脑部核磁共振成像检查。结果显示，重度电视人群的大脑结构出现了缩小，其中男性的影响更显著。

Mozilla 工程总监 Sylvestre Ledru 通过 [email protected] 邮件列表宣布，从 2026 年 9 月起 Firefox 桌面版和 Android 版本的发布周期从 4 周减少到 2 周。他表示这是一次实验，并不意味着工作进度加倍，并不会仓促交付新功能。当前的目标是在 9 月 1 日释出 Firefox 155 而不是原计划的 9 月 15 日，“我们将密切关注这一变化在实际中的效果，根据需要进行调整。”Firefox 接下来的两个版本 Firefox 153（ESR） 和 154 仍然是每隔 4 周释出，Firefox 155 开始提前两周释出。

世界各地的亚马逊 AWS 客户周五可能都心跳加速，他们看到了远远超过他们想象的账单数字，很多每月只花几美元的客户看到了数亿乃至数十亿美元的账单，很多企业甚至看到了数以万亿美元的账单。亚马逊证实其计费系统出现问题，账单预估数字不正确，在问题解决期间它暂停了账单更新。亚马逊称它已经识别问题是单位导致的，所有受影响客户预计到 7 月 19 日 12:00 AM PDT 将会完全恢复正常。亚马逊没有详细解释单位问题，猜测是原来按 GB 收费的系统漏掉了 GB，而系统默认按 Byte 计费，1 GB = 1,073,741,824 Bytes，这意味着一小时内账单费用就会膨胀十亿倍。

危地马拉舒尔通玛雅遗址一面墙壁上镌刻的数学公式，让学界首次确认了一位重要的玛雅数学家兼天文学家的真实姓名。研究团队指出，这位名为 Sak Tahn Waax（意为“白胸狐狸”）的学者足以比肩人类历史上的数学巨匠。 墙壁绘有人物画像与象形文字，其中包含基于天文历法的精密数学运算。玛雅人曾依靠这些运算确定国王登基等重大仪式的举办时间。研究团队重点分析了编号为“19号文本”的一组象形文字。“19号文本”是一组呈L形排布的11个象形文字，整体高度约10厘米。研究发现，前 9 个象形文字完整记录了玛雅历法与天文周期的换算逻辑。这套公式破解了一个2920 天周期的拆分规律，可适配玛雅各类历法单位。2920 天是玛雅文明的核心天文周期，完美契合5个金星周期（每周期584天）与8个太阳年（每周期365天）。不仅如此，该运算还将2920天与乌伊纳尔（每月20日）、卓尔金历（260天神圣历法）、通年（每年360天）以及780天火星周期建立了精准数理关联。 团队在“19号文本”倒数第二个符号中破译出“如是说”的句式，其后紧跟的最后一个象形文字便是署名 Sak Tahn Waax，代表该学者是这套运算公式的创作者。从铭文缺失女性专属前缀可判定，这是一位男性学者。

天文学家直接探测到系外行星 LHS 1140 b 大气层中的氦气。LHS 1140 b 是一颗岩石行星，距离地球 48 光年，位于其母星的宜居带。这是首次在宜居带内的类地球岩石行星确认探测到大气层。LHS 1140 b 围绕着一颗比太阳更小更冷的红矮星运行，其轨道比日地距离更近，温度适宜，其表面可能存在液体水，可能有铁质内核。就对地球的生命研究而言，液态水是生命存在的必要条件。论文第一作者 Collin Cherubim 表示需要进一步研究去确认行星是否存在水。

在今天的意大利漫步，我们仍然能看到有近两千年历史的混凝土建筑。相比下，现代混凝土建筑在百年内就会坍塌。为什么罗马时代的混凝土耐用性如此好？科学家认为要归功于名为“火山灰反应”的关键化学过程——即火山灰与石灰和水发生反应。根据发表在《Science Advances》上的一项研究，名为碳化的反应也有助于增强混凝土的耐久性。研究人员从有 1900 年历史的哈德良(Hadrian)皇帝庄园的马桶座圈上收集了混凝土样本，用高倍显微镜观察，用 X 射线扫描分析其化学成分。研究发现，样本含有火山灰、石灰和水等材料的证据，但对混凝土孔隙和裂缝的观察显示，方解石是其主要的粘结剂。当大气中的二氧化碳与混凝土中的钙化合物发生反应时，会形成坚硬的方解石矿物，它含有大量的碳酸钙。方解石填充了混凝土中的细小裂缝和孔隙，使古老的建筑结构能随着时间的推移而加固和修复。

黑胶唱片过去十年一直是实体音乐复兴的代表，但今年上半年美国 CD 唱片销量涨幅远超黑胶唱片。CD 销量飙升 16% 达到 1630 万张，相比下黑胶唱片销量上涨 2.4%。韩国 K-pop BTS 热门专辑《ARIRANG》在 CD 销量增长中起到了重要作用，但排除 K-pop 之后，CD 销量仍然同比增长了 6.7%。美国包括 LP、CD 和磁带的实体专辑总销量上半年增长了 7.8% 达到 3820 万。实体音乐复兴的一大原因是年轻一代听歌习惯发生了变化，六成 Z 世代听众表示最常听的是音乐是 1990 年代或更早期的，相比下 2021 年这一比例仅为 18%。无论是怀旧、通过流媒体发现新音乐，还是想要拥有艺术家作品的实体唱片，年轻听众在拥抱实体唱片。

研究团队结合了地面大型巡天计划 LAMOST DR7 的光谱观测，与 ESA 盖亚太空望远镜的高精度视差资料，建立起包含 155,600 颗邻近太阳次巨星（Subgiant stars）的庞大基准样本。由于这类恒星形成于银河系早期，其化学组成几乎保留了最原始的样貌，是推算时间极佳的恒星化石。团队利用马可夫链蒙地卡罗（MCMC）算法重建真实年龄分布，推导出样本中最古老恒星的真实年龄为 137.3（+1.8 / -1.5）亿年，若考虑大爆炸后约 2 亿年才形成首批长寿命恒星，这与宇宙微波背景辐射预测的 138 亿年宇宙年龄完全相符。最新结果显示，目前并没有可信的观测证据支持宇宙年龄超过 138 亿年。

微软宣布开源 Comic Chat，源代码采用 MIT 许可证托管在 GitHub 上。Comic Chat 是一款能自动将 IRC 中的对话转换为漫画格形式的聊天客户端，其中包含插图人物、对话气泡和表情。它的一个目的是帮助世界认识 Comic Sans 字体。Comic Sans 字体最早由微软字体设计师 Vincent Connare 于 1994 年设计，在 Comic Chat 中找到了它真正的家。它非正式的手写风格与软件的对话气泡完美匹配。Comic Chat 最早于 1995 年开发，1996 年随 Internet Explorer 3 推出。开源 Comic Chat 的一个意图是保存软件历史，以及让开源社区在此基础上探索、学习和二次开发。

月之暗面宣布了 2.8 万亿参数开放权重模型 Kimi K3，完整的权重将于 7 月 27 日发布。月之暗面称，Kimi K3基于 Kimi Delta Attention 和 Attention Residuals 构建，参数规模 2.8 万亿，具备原生视觉功能和 100 万个词元上下文窗口。它是全球首个 3 万亿参数级别的开放权重模型，专为长程编码、知识工作和推理等前沿智能领域而设计。在基准测试中，Kimi K3 整体性能仅落后于 Claude Fable 5 和 GPT-5.6 Sol。Kimi K3 现已在 Kimi.com、Kimi Work、Kimi Code 和 Kimi API 上线。发布初期 Kimi K3 将默认采用最大思考强度模式，低强度和高强度模式将在后续更新中推出。