AI 日報

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：The following article originally appeared on Venkatesh Rao’s Substack, Contraptions, and is being republished here with the author’s permission. The sole athletic achievement of my life came in 1993: winning the IIT Bombay freshman 50m freestyle race with a time of 41s. That got me into the college swim team (it was a bad recruitment […]

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Reactiv used Amazon Bedrock AgentCore to build a multi-agent AI Scheduler that autonomously refreshes Shopify merchants' mobile apps on a schedule, reducing merchant configuration time by 80% and getting to production 33% faster.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Concurrency sweeps help you right-size a generative AI endpoint on Amazon SageMaker AI by systematically benchmarking it at increasing load levels. This post walks through deploying a model, running automated concurrency sweeps with the CreateAIBenchmarkJob API, and using the results to make data-driven capacity decisions about fleet size.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：In about four weeks, Trane Technologies built an AI-powered agentic solution on Amazon Bedrock AgentCore that reduced a 20-minute, multi-screen building diagnostic workflow to a 20-second natural language interaction, a 60x improvement in time-to-insight. This post shares the architectural approach and key design decisions behind the solution.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Learn how Tata Elxsi built IRIS, a real-time industrial safety platform on AWS. IRIS filters camera video at the edge, streams metadata through Amazon Kinesis, runs computer vision on Amazon SageMaker AI, and correlates detections into high-confidence alerts, detecting unsafe conditions in seconds instead of minutes.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Public sector agencies process large volumes of unstructured evidence, such as body camera footage and scanned documents. This post shows how to combine Amazon Bedrock Data Automation with the Model Context Protocol (MCP) to turn that data into structured insights and surface them through natural language queries in Salesforce Agentforce.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Have you ever read a paper in Science or Nature and thought, “Man, that research was so cool. I wish I could try that method on my own data”—only to spend a week wrestling with someone else’s undocumented repo, broken dependencies, and half-finished readme.txt? Well, now you can, more or less. Say hello to Paper2Agent, a new open-source framework that transforms academic reports into interactive AI agents you can talk to. Give it a paper along with the accompanying codebase, data or other supplementary material, and the system automatically extracts the core workflows, then spins up a tested, runnable toolkit that you can use on your own datasets. The concept may sound a little like Google’s NotebookLM (now called Gemini Notebook), which lets you upload documen…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Alongside warnings about its ability to destroy humanity, artificial intelligence is also hailed for its potential to revolutionise health treatment. Just how far can it take us? Her name was June. June would be the last month she ever knew. I remember the way her bed made her so small, swallowed her up in the middle of the room, turned as it was to face the window. With her liver failing, her skin was a ghostly apricot. June – I called her Nan – had been stalked by the king of terrors, the emperor of all maladies: cancer. That son of a bitch. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Every AI agent is only as good as the context it is grounded in. Ask an agent a question...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：To build and deploy sophisticated robotics applications that can perceive, reason and act in dynamic environments, developers need new physical AI models and tools. The ROS open framework is a project from Open Robotics that helps humans build robots. NVIDIA Isaac ROS 5.0 — a collection of GPU-accelerated packages built on ROS, released today at […]

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：The zero-day exploit required local access to the user’s device, but gave potential attackers access to Muse accounts. | Image: The Verge Meta has issued a patch for its Muse macOS app following the discovery of a zero-day vulnerability that could allow someone to take control of the AI agent. The bug found by security researcher Patrick Wardle utilized an undocumented Muse setting that enabled potential attackers running local code to redirect transcription processing from Meta's servers to their own endpoint, Ars Technica reports, giving the attacker access to the Muse account. Several design decisions reportedly enabled this flaw, including having Muse dictation occur in the cloud instead of on-device, and allowing any app to control all of Muse's undocument…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：NVIDIA researchers have released SoL-Pi, 4 harness mechanisms for the open-source Pi coding agent, discovered by an AI running auto-research loops across 535 environments. On EdgeBench, SoL-Pi cuts token traffic by 44.7% to 49.0% and API cost by roughly 33%, while keeping about 94% of Pi's score on GPT-5.6 Sol and Opus 5. The post NVIDIA Introduces SoL-Pi: Auto-Research Loops That Cut Coding Agent Token Traffic by Up to 49% appeared first on MarkTechPost.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22104v1 Announce Type: new Abstract: As automated scientific discovery advances, Large Language Models (LLMs) can now generate research ideas at an unprecedented scale, shifting the bottleneck from idea generation to idea evaluation. Existing evaluators mainly rely on parametric LLM knowledge or unstructured retrieval, producing judgments that lack the experience-grounded reasoning used by human instructors. To address this, we propose DeepInstructor, an agentic framework that formulates idea evaluation as reasoning over structured scholarly experience. DeepInstructor constructs an Experience Graph from 58,607 peer reviews and employs a ReAct-based agent to retrieve dimension-specific evidence for traceable evaluation. We further introduce DeepInstru…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22090v1 Announce Type: new Abstract: An LLM producing the response pattern associated with a human psychological effect is not the same claim as the LLM possessing that bias. We present PsyAgentBench, a benchmark that re-runs classic psychology experiments on LLM agents under a factorial design built to separate these: each paradigm is run with the paradigm explicitly labeled in the prompt (named) or framed as a routine task (blind), and on the literal textbook version of the task (canonical) or a structurally matched variant written to reduce lexical and scenario overlap with likely training data (counterfactual), crossed with a persona manipulation. Across five completed paradigms, evaluated on up to three open-weight model families with 41,904 tri…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22120v1 Announce Type: new Abstract: Test-time self-evolving agents improve by reusing past experience, yet sparse-reward trajectories contain failures, loops, and detours, while summaries often omit the state conditions and action dependencies needed for execution. We study executable Walkthrough induction from sparse-reward trajectories: extracting compact, state-conditioned, and verifiable procedures. Our key observation is that delayed credit identifies actions associated with progress but cannot determine whether they produce facts required by later actions. We propose Trace, a credit-guided, dependency-grounded framework that compiles noisy trajectories into executable Walkthrough Memory. It detects progress anchors from rewards and persistent…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Learn how to use Engram effectively, from writing topic descriptions that control extraction to choosing bounded topics and scopes, selecting a retrieval mode, and placing memories in your prompt without hurting prompt-cache efficiency.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯： Cloudflare Python Workers are now generally available After a two year preview, Cloudflare's support for running Python code in their server-side Workers platform is now stable: "Python is now a first-class, fully supported language on the Cloudflare Developer Platform". A neat thing about this is how it works. Cloudflare are running Python compiled to WebAssembly via Pyodide in their V8-based workerd runtime. This comes with some limitations, documented here - most notably both multiprocessing and threading are non-functional in the WebAssembly VM. One particularly interesting detail of this is the local development environment story - their pywrangler development tool (confusingly packaged as workers-py on PyPI) runs a full local simulation of their stack, i…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：The definitive Jev podcast with its lead creator.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Many developers find that an agent idea works inside Claude Code or Codex, then struggles once they rebuild it with their own loop. The Strands Agents team at AWS is targeting that gap with Strands harness, a fully assembled, general-purpose agent harness. It runs locally or deploys to a cloud provider, ships for Python and […] The post AWS Strands Agents Team Releases Strands Harness: An Open-Source Agent Harness With 28% Lower Token Cost at Comparable Accuracy appeared first on MarkTechPost.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Amazon Web Services (AWS) is lifting the lid on a new open source, general-purpose AI agent, designed to give developers The post AWS open-sources an AI agent it says is 45% cheaper than Claude Code and Codex appeared first on The New Stack.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Use Jev as a judge for LangSmith evals to evaluate agent traces with faster, cheaper structured feedback across production runs, datasets, and regression tests.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：BMW Group operates CLEA, a FinOps platform monitoring more than 14,000 cloud accounts. This post shows how BMW added automated daily cost anomaly detection, moving from reactive dashboards to proactive alerts using Prophet forecasting, AWS Step Functions, and a serverless pipeline that processes every account for about $50 per month.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Positron, Posit's IDE for data science, now runs on Amazon SageMaker AI. This post shows how a data scientist explores an Amazon Athena table, validates features in R, trains an XGBoost model in Python, deploys a real-time SageMaker AI endpoint, and reports results with Quarto, all in one governed SageMaker Studio Space.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Learn how Benchling built a defense-in-depth security architecture to run untrusted, AI agent-generated scientific code across thousands of life sciences tenants using Amazon Bedrock AgentCore Code Interpreter in VPC mode, combined with Amazon Route 53 Resolver DNS Firewall and VPC endpoint policies to block data exfiltration, including through DNS.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Physical AI is moving rapidly from research to large-scale deployment. By 2035, ABI Research projects an installed base of 49 million level 3-5 autonomous vehicles (AVs), while Omdia estimates that roughly 60 million industrial robots will be deployed between 2026 and 2035. As these machines enter roads, factories, warehouses and other environments shared with people, […]

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Today, Egypt’s AI builders gathered in the Grand Egyptian Museum for a reception that highlighted the nation’s rapidly growing AI ecosystem — spanning AI natives, developers, researchers, startups and enterprises — building applications across industries. The event included a keynote from Paolo Guglielmini, vice president of EMEA at NVIDIA. Ahmed Mostafa, regional AI adoption lead […]

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Phylo brings frontier AI to more scientists with open models on Fireworks Join us for our inaugural conference, Forge 2026 Blog Phylo Brings Frontier AI To More Scientists With Open Models On Fireworks Phylo brings fron…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯： Release: llm-typesafe 0.1a0 I built this new plugin for LLM to add support for TypeSafe AI's new Jev model. Install it like this: llm install llm-typesafe Then set an API key (get one here, the waitlist seems to move pretty fast): llm keys set typesafe # Paste key And now you can ask yes/no "noul" questions like this: llm -m jev 'Please refund my last payment.' \ -s 'Does this message explicitly request a refund?' Output: {"type": "noul", "noul": 0.99} Or choice questions like this: cat message.txt | llm -m jev \ -s 'Which team should handle this message? If billing and technical issues both occur, choose billing.' \ -o answer_type choice \ -o criteria '{ "billing":"Charges, invoices, payments, or refunds", "technical":"Problems installing or using the product…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：In this article, you will learn what embedding drift is, why it matters for production large language models, and how to implement two practical techniques...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：GPT‑6 Astra allowed Parallel’s agents to research and synthesize labor-market data in half the time and at half the cost vs. prior models.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：The early rapid expansion of AI capabilities that focused on frontier models was largely ushered into the world by a few powerful, US-based AI labs. Open-weight models released from labs in China, early on from DeepSeek, and later from Moonshot, Z.ai, and others, have in part disrupted that dominance. But growing concerns about the concentration […]

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Inside the RSI post-training pipeline

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Burnham is heading to New York for the United Nations general assembly Kanishka Narayan, the minister for AI, is on the trip to New York with the PM. In a video that he recorded on the plane, he says the fact that he is going shows that AI is now “top of the international agenda”. He says his message to the industry has been “take risks seriously”. According to the overnight briefing from No 10 ahead of his speech to the UN, Andy Burnham will make AI (artificial intelligence) one of the main themes of his visit to New York. Downing Street says: Major global challenges such as AI, the conflict in Ukraine and security in the Middle East are expected to dominate the agenda of the prime minister’s meetings in New York over the next 24 hours, as world leaders conver…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：crowning a new Chinese frontier lab

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：SpaceXAI has released Grok 4.7, its new flagship model for coding, agentic tasks, and knowledge work. Grok 4.7 is built on a larger base model and a longer reinforcement learning run. It still ships at the same price and speed as Grok 4.6. Is it deployable? Yes, as a hosted model. You can call grok-4.7 […] The post SpaceXAI Releases Grok 4.7: A Larger Base Model at the Same $2/$6 Price as Grok 4.6 appeared first on MarkTechPost.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22325v1 Announce Type: new Abstract: Large language models have shown considerable potential for natural-language-driven parametric CAD modeling. However, a fundamental contradiction exists between their probabilistic generation and the deterministic requirements of CAD modeling, resulting in limitations in reliability, design-intent preservation, and geometric validity. Existing methods typically rely on large-scale annotated datasets, lack explicit modeling of design intent, and underutilize the deterministic capabilities of CAD kernels. To address these limitations, we propose ReliCAD, a unified framework that transforms uncertain LLM generation into reliable parametric CAD modeling. Through explicit design-intent modeling, ReliCAD converts user r…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22319v1 Announce Type: new Abstract: The rapid progress of artificial intelligence is reshaping robotics and accelerating the adoption of learning-based approaches. While purely data-driven methods have achieved remarkable success in computer vision and natural language processing, robotics remains constrained by limited data, complex real-world interactions, and the need for reliable operation. These challenges have motivated the exploration of physics-embedded robot learning, which embeds physics priors into learning algorithms. By encoding the underlying physical laws and constraints, physics priors can complement limited data with robotics-specific inductive biases, potentially improving generalization, interpretability, and sample efficiency. Ho…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22293v1 Announce Type: new Abstract: Vision-language models (VLMs) and vision-language-action models (VLAs) are increasingly deployed in real-world applications. There, a small perturbation to the recorded camera image may change a decision significantly. However, existing benchmarks for these models only sample perturbations, which does not guarantee the absence of a failure in the untested region. We present the first robustness validation of six VLMs (drawn from the Gemma, InternVL, LLaVA, and Qwen families) and five VLAs (drawn from the GR00T, OpenVLA, and $\pi$ families) over entire continuous regions of photometric and geometric image perturbation: brightness shifts, camera rotations, and their composition. To this end, we build on the validati…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22291v1 Announce Type: new Abstract: This paper presents a comprehensive experimental evaluation and detailed analysis of state-of-the-art multi-object tracking algorithms, with an emphasis on quantifying the individual contributions of detection and association components to overall tracking performance. Unlike existing surveys that primarily offer theoretical categorizations or taxonomies of tracking methods, our work adopts a rigorous experimental perspective grounded in publicly available implementations, providing practical guidance for researchers and practitioners in method selection and system design. We introduce a unified pipeline diagram that consolidates the core components across the two main branches of visual multi-object tracking: tra…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22283v1 Announce Type: new Abstract: Understanding the design space of streaming video diffusion is essential to exploring its potential for generation quality and computational efficiency. We develop a unified analytical framework that relates model and sampler choices, historical conditioning, execution scheduling, and training strategies. The framework accommodates a broad family of causal context-selection policies and makes their computational dependencies and training-inference alignment explicit. Within this design space, we study three representative policies: clean, same-level, and progressive history. On the full VBench prompt set, same-level and progressive history achieve aggregate scores of 85.24 and 85.60, respectively, compared with 84…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22282v1 Announce Type: new Abstract: Reconstructing visual stimuli from electroencephalography (EEG) is difficult because scalp measurements have high temporal but limited spatial resolution, and paired EEG-image datasets remain small relative to modern generative-model training corpora. We present a reproducible single-subject baseline on THINGS-EEG2 that first tests the more defensible question of whether EEG can retrieve the viewed stimulus in a visual embedding space. A compact temporal-spatial convolutional encoder maps repetition-averaged EEG (63 by 250) to provided 512-dimensional ViT-B/32 image features. Model selection uses a concept-disjoint validation split, and final evaluation uses the official 200-image, 200-concept test gallery. Across…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22281v1 Announce Type: new Abstract: Foundation models have recently demonstrated strong capabilities across a wide range of medical imaging tasks. However, their performance in structured clinical interpretation settings remains insufficiently explored. In lung cancer screening, interpretative variability persists despite standardized frameworks such as Lung-RADS. In this study, we evaluate MedGemma, a medical general-purpose foundation model derived from Gemini and its fine-tuned version adapted for lung cancer detection and diagnosis, compared against radiologists performing Lung-RADS v2022 assessment on the NLST dataset. Twelve radiologists independently evaluated each case in a multi-reader design, enabling quantification of inter-reader variabi…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22272v1 Announce Type: new Abstract: We formulate \emph{Artistic Intelligence} as exploration driven world realization, leaving space for creative possibility while preserving the semantic, artistic, and compositional structure that must remain true. Moonworks Lunara, a text-to-image model, implements this framework with a novel Diffusion Mixture Transformer architecture. A new training algorithm iteratively evolves the data distribution through informative sample acquisition and targeted injection of human-created art. We benchmark Lunara against seven image-generation models, including FLUX.2-Klein-4B, Qwen-Image (20B), and GPT-Image-1-Mini. With GPT-5.6 Sol as evaluator, Lunara ranks first in \emph{Aesthetic Quality (8.473 vs. 8.457 GPT-Image-1-mi…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22271v1 Announce Type: new Abstract: Multimodal stroke recurrence prediction requires effective integration of heterogeneous clinical and imaging data, yet modality imbalance often causes models to over-rely on dominant modalities and underutilize complementary information. While self-supervised pretraining and selective parameter freezing are commonly employed to improve representation learning and fine-tuning stability, their effect on modality contributions and cross-modal behavior in multimodal medical models remains largely unexplored. In this work, we investigate whether image pretraining on 3D CTA scans reduces modality imbalance and improves cross-modal integration for stroke recurrence prediction, a clinically critical task we recently addre…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22101v1 Announce Type: new Abstract: Large language models can process increasingly long prompts, yet their ability to locate and use decisive evidence may degrade as irrelevant or confusable context is added. We formulate this phenomenon, which we call context poisoning, as extreme-value interference in attention: the decisive-evidence score is upper-bounded, while the maximum score among effective distractors grows with their number. Under a softmax retrieval abstraction, we derive a finite-sample upper bound showing that maintaining a fixed accuracy target above base rate requires the evidence margin to scale as $\Omega(\sqrt{\log N})$, where N denotes the effective distractor count rather than necessarily the raw context length. The analysis conn…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22100v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) improves language models with retrieved evidence, but processing many long passages is costly and can introduce distracting information. Soft compression addresses this challenge by encoding passages as compact sequences of continuous memory embeddings before generation. However, existing methods typically assign each retained passage an identical number of memory embeddings, irrespective of its query-specific relevance. To address this, we propose AdaMem, a relevance-guided soft-compression framework that maps learned passage-relevance estimates to a query-dependent allocation of a fixed memory-token budget. A shared query-conditioned compressor produces both continuous passag…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22099v1 Announce Type: new Abstract: Cooking is a complex process that transforms raw ingredients into delicious and nutritious dishes, yet the recipes that encode this process remain largely free text; readable by people but not directly computable. Existing recipe collections capture fragments of this information, but no shared representation links a recipe's structured ingredient composition, its geo-cultural provenance, and its nutritional profile within a single queryable schema. We address this representation gap by formalizing a framework for recipe data structure that decomposes each recipe into typed ingredient entities, grounds those entities in a reference nutritional database, and annotates them with geo-cultural and dietary context. We p…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22098v1 Announce Type: new Abstract: Speculative decoding accelerates language-model inference by letting a cheap drafter propose tokens that the target model verifies in parallel. Recent block drafters make drafting nearly free: a single backbone pass emits an entire block of draft tokens. Draft trees promise a further gain -- several alternative continuations verified in one target forward -- but existing constructions rank candidates by per-position marginals that ignore which parent a candidate extends, so on semi-autoregressive drafters wider trees mostly add mis-ranked nodes; and a tree of fixed size ignores how much speculation each decoding round, and each serving load, can support. We introduce TreeSpark, which reads a parent-conditioned dis…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22097v1 Announce Type: new Abstract: The evaluation of large language models (LLMs) on coding tasks has primarily focused on performance metrics such as pass@k. As LLMs continue to advance, many models now meet baseline performance requirements, reducing the discriminative power of performance-based evaluation alone. Yet a key question remains largely unexplored: how do LLMs differ in their coding behavior? We propose CLIC (Code Learning for Identification and Comparison), a visual analytics approach that characterizes LLM coding behavior through token-frequency analysis. CLIC represents each code sample as a feature vector of token frequencies and trains an interpretable decision tree to separate two LLMs' code sets. Beyond classification accuracy,…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22094v1 Announce Type: new Abstract: Content moderation systems traditionally entangle multimodal understanding with policy-specific classification, requiring full pipeline retraining for every policy change and suffering from label scarcity since multimedia cannot be meaningfully augmented. We propose Summarize-Judge-Refine (SJR), a two-model architecture that decouples these concerns via a natural language interface: a multimodal Content Model produces structured text summaries, and a text-only Policy Model classifies them against policy definitions. An iterative co-training loop refines the Content Model via GRPO to produce policy-relevant summaries, while text-space augmentation generates adversarial summary variants---an augmentation pathway imp…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22091v1 Announce Type: new Abstract: Retrieval over a personal memory store is retrospective: it surfaces what resembles the query, and it is blind to what the user has committed to do. We describe a prospective term for memory retrieval that costs no inference at query time. Commitments are held in an explicit ledger as dated or trigger-conditioned entries; memory items linked to a firing entry receive a salience boost, blended multiplicatively into embedding-based retrieval so that relevance remains sovereign. On a synthetic prospective-memory task set modeled on TriggerBench's published structure (48 blind-authored dialogues, 175 tasks), the term raised recall@5 on the hard stratum from 0.000 to 0.955 at the default blend weight and to 1.000 under…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22121v1 Announce Type: new Abstract: Passively collected mobile phone location data provide large-scale, longitudinal observations of human mobility but do not directly reveal activity purposes. The functional characteristics of visited locations offer useful contextual information, yet their relationship with activity purpose remains uncertain, particularly in mixed-use urban environments. We conceptualise activity pattern mining as an integrated process of representation, clustering, and interpretation, and propose the Activity Chain Encoder (ACE) for the representation stage. ACE is a self-supervised model that combines pre-trained urban embeddings with visit timing and duration and uses a Transformer to model the sequential organisation of stays.…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22117v1 Announce Type: new Abstract: Location representations provide mobility models with fundamental information about the spatial position, functional characteristics, and relationships of places. However, existing embeddings are often dependent on mobility observations, unable to represent unseen locations, and weakly constrained to retain geographic distance. This limits their reuse across datasets and mobility tasks. To address these limitations, we propose LE4Mob, an inductive, distance-aware, and geography-derived location embedding framework for mobility modelling. LE4Mob extends contrastive language-location pre-training while introducing a distance-aware regularisation objective that encourages the embedding space to preserve spatial relat…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22115v1 Announce Type: new Abstract: Zeroth-order optimization (ZOO) estimates updates from function evaluations, making perturbation queries a primary cost. Fixed budgets spend the same number of queries at every step, while adaptive controllers may offset their savings by using additional oracle calls to test estimator reliability. We introduce ZoAQ, an adaptive ZOO method built around query reuse. Rather than discarding past evaluations after each step, ZoAQ makes them useful for both the next update and the decision to query further. This enables adaptive query allocation without extra validation queries. Our analysis characterizes when this agreement identifies an update that supports descent and guides the controller to a sufficient query budge…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22109v1 Announce Type: new Abstract: Selective on-policy distillation trains a student only at the token positions a selector scores highest, and the literature compares selectors under a single shared learning rate--a control chosen to be neutral. We show it is not. Under LoRA on GSM8K (Qwen2.5-1.5B student, 7B teacher), across an 8x learning-rate grid, dense supervision is statistically flat (swing 1.8 pp, p=0.26) while every selective arm moves with the rate: 5.4 pp for a random 5% subset, 6.7 pp for a total-variation selector, up to 17.7 pp for a teachability selector. Consequently the dense-versus-selective verdict reads 10.1 pp at lr=1e-4 but 5.1 pp at 5e-5--a 2.0x difference decided by a parameter the protocol treats as scenery--and two of six…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22107v1 Announce Type: new Abstract: Making prediction with multimodal data is widely used in diverse scenarios. Existing multimodal fusion models, once deployed, can only handle predefined modalities (e.g., vision, text and audio) and single tasks, making it difficult to quickly adapt to new downstream applications. Therefore, a natural yet rather aggressive question arises, whether there exists a general multimodal fusion model that can be applied to arbitrary modality combinations and arbitrary prediction tasks. We argue that a unified multimodal fusion model should not depend on specific modalities and instead encode transferable patterns of multimodal correlation. To this end, we propose a simple and effective learning paradigm based on training…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22106v1 Announce Type: new Abstract: Accuracy of Low-bit quantization of linear layers is often dominated by a small number of outliers. Although existing methods, such as smoothing, rotation, or residual-based approaches, may mitigate this problem, they often introduce new accuracy bottlenecks to weights. Besides, most of these techniques are implemented as online approaches, which can result in heavy execution overheads. To address the afore-mentioned issues, We propose PRQuant (Permutation Residual Quantization), a training-free and low-overhead framework that combines channel reorganization with static weight-side residual compensation. After AWQ-style scaling, PRQuant identifies the input channels that contribute most to weight quantization erro…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：OpenAI outlines priorities and principles for rigorous, secure, and independent third-party AI safety assessments of frontier models and safeguards.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯： Last week TypeSafe AI unveiled Jev, their first example of a new category of model that they are calling "System One models" (I'm with Maggie Appleton, I think "decision models" is a better name for these). Jev is an interesting variant on the usual LLM format: it still accepts text inputs, but instead of text output it returns floating point numbers corresponding to categories, yes/no questions, ratings, and associated confidence scores. TypeSafe describe Jev like this: Think of Jev as a frontier-intelligence function call: unstructured state in, typed probabilistic decisions out. It's also very fast, and really cheap. Regular LLMs are priced in terms of input and output tokens, with output generally charged at significantly higher rates. Jev charges only for…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：xAI's Grok 4.6 is now available in Amazon Bedrock: a frontier model for long-running agents, coding, and knowledge work, with a 500K token context window and four reasoning effort levels. It runs on both the bedrock-mantle and bedrock-runtime endpoints, with Converse API and cross-Region inference support.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Alibaba's Qwen team has released Qwen-Image-2.1, a 7B diffusion transformer that handles text-to-image generation, multi-reference editing, and native RGBA transparency in one checkpoint. A prefix KV cache speeds up edits with up to 10 reference images. The weights are public, but commercial use requires a separate license from Qwen. The post Alibaba Qwen Releases Qwen-Image-2.1: A 7B Open-Weight Model for Image Generation and Editing appeared first on MarkTechPost.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：EXL built an AI-powered Medical intelligent document processing (IDP) solution on AWS, combining IDP with domain-specific large language models on Amazon SageMaker and Amazon Bedrock to extract, summarize, and query medical records at enterprise scale and cut claims review time from over 100 minutes per case.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：António Guterres says humanity faces unprecedented convergence of threats but mechanisms for collective action are under growing strain The world’s fault lines have turned from cracks into canyons as superpowers show that they alone do not have the military, economic and technological power to guarantee security, or solve the world’s problems, the outgoing secretary general of the United Nations has said in his final address to the general assembly. In a speech laced with despair but also glimpses of defiant hope, António Guterres warned that over the last decade, largely spanning his eight-year period in office “wars erupted with devastating consequences and dragged on with despicable cruelty. Civilians were targeted. Human rights trampled. Hunger weaponised.…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Former chancellor, now senior figure at OpenAI, says UK needs datacentres to maintain ‘sovereignty’ over AI tech UK politics live – latest updates Datacentre nimbys are holding Britain back, George Osborne has said in reaction to nationwide protests about the giant water-and-energy guzzling structures. The Tory former chancellor is now the head of AI for countries at OpenAI, an artificial intelligence company and the developer of ChatGPT. His job is to represent the company to governments around the world, and he also has a role in enabling the building of datacentres and other infrastructure. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：It’s becoming obvious to many Americans that corporations have been raking it in while workers take it on the chin Last week, the unlikely duo of Bernie Sanders and Steve Bannon both spoke before a Washington group demanding a stop to AI and its datacenters. Notably, both lambasted the corporate oligarchs behind AI. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：RBA governor says AI is ‘great white hope’ but its adoption is so far adding to inflation, rather than economic growth The Reserve Bank governor said AI could be a bubble and there was no evidence it is making the economy more efficient, as the Albanese government claims the technology will solve Australia’s economic malaise. Ahead of an expected interest rate rise next week, Michele Bullock also said the slump in house prices was deeper than others in Australia’s recent history. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Canadian province alleges deadly attack could have been prevented if company had warned police of shooter’s ChatGPT use British Columbia has sued OpenAI in California, saying a mass shooting at a school in the province could have been prevented if the company had warned local ⁠law enforcement that the shooter ⁠had used ChatGPT to ​plan the massacre. The lawsuit filed in San Francisco federal court on Monday names OpenAI and its CEO, Sam Altman, as defendants. It is seeking damages to fund recovery efforts in the province after the February attack, as well as an order directing changes to the way the company handles ChatGPT conversations that could lead to violence. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：A hard model swap exposes every user at once, and rolling back means cold-starting the old deployment under pressure. Here's how staged traffic ramps, metric gates, and automatic rollback work on dedicated inference.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：At Lib Dems’ conference in Brighton, leader will accuse Andy Burnham of relying on Donald Trump and ‘tech bros’ to solve problem Ed Davey is to call for a global nuclear-style non-proliferation treaty to halt the development of super-intelligent AI when he makes his keynote speech at the Liberal Democrat conference, accusing Andy Burnham of relying on Donald Trump and “tech bros” to solve the problem. When he addresses the faithful on the final day of the gathering in Brighton, the party leader will call for a global pause on super-intelligent AI, comparing it to a nuclear arms race that could result in technology that could “destroy us all”. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：California Gov. Gavin Newsom has signed seven bills designed to prevent AI data centers from passing utility costs onto residents, as reported earlier by the Los Angeles Times. The package of laws requires the California Public Utilities Commission to introduce a new rate classification for data centers, while forcing them to pay for upgrades to local power grids and water systems. Other bills included in the package will require proposed data centers to disclose their estimated water use to local governments, along with information about energy efficiency and drought planning. They must also meet certain energy, water, and fuel consumption … Read the full story at The Verge.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Every AI factory needs power and cooling that fit its computing architecture. As AI infrastructure expands, power, cooling, water, site and grid constraints are shaping what builders can deploy. Choosing products that fit the complete factory design helps builders turn computing capacity into useful AI output. To help builders make those decisions, NVIDIA is introducing […]

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discussion | Link

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：The former UK deputy prime minister is a board member of the UK-based company and owns 917,000 shares Nick Clegg could make approximately $40m (£30m) from the planned float of Nscale as the UK-based datacentre company prepares for a US stock market listing. The former UK deputy prime minister owns more than 917,000 shares in the startup, which is reportedly seeking a valuation of up to $35bn in a flotation. Based on Clegg’s stake of 0.12%, that could yield a valuation of about $42m for the ex-Liberal Democrat party leader. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Podcast #19

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：A new AI control system lets MIT’s tiny flying robot move with insect-like agility, boosting its speed by about 450 percent and allowing it to pull off 10 somersaults in 11 seconds. The technology could eventually enable miniature robots to search earthquake rubble and navigate dangerous spaces that conventional drones cannot reach.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：As animal behaviorists increasingly use AI in their research, they grapple with how to wield the technology responsibly I’m sitting in a small boat in Shark Bay, Australia, eavesdropping on a conversation between some of Earth’s most intelligent beings. The motor is off, the turquoise sea is glassy, and Stephanie King and I are motionless, transfixed by the whistles, clicks and squeezy, whirring noises made by the dolphins swimming below us. If anyone could tell me what these big-brained social mammals are saying, it would be her. Based at the University of Bristol, she co-leads Shark Bay Dolphin Research, one of the longest-running wild dolphin research projects in the world. She has spent thousands of hours watching and listening to this population. Continue…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22317v1 Announce Type: new Abstract: World models support autonomous driving by predicting the scene evolution associated with candidate trajectories. Driving dynamics differ in predictability, motivating a distinction between regular evolution and event-induced deviations that call for selective correction. We propose EditWM, a world model that decomposes future prediction into normal evolution and event-driven incremental correction in compact visual feature space. A trajectory-conditioned normal predictor provides the base forecast and is then frozen for correction learning. A correction decoder compares this forecast with observation history and planned actions, producing a bounded feature update whose contribution is regulated by a learned gate.…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22299v1 Announce Type: new Abstract: When a robot faces unfamiliar physical conditions, a common approach is to collect evidence about what changed and adapt. For such diagnosis to improve behavior, six ordered empirical conditions must hold: a meaningful reference, identifiability of the physical condition, use of the acquired evidence, decision value, selection value over a fixed alternative, and safe realization. We test this chain in controlled and public environments. It holds end to end in our controlled environments. After transfer to unseen mechanisms, however, it breaks at evidence use. On decisions requiring the full trace, the frozen decoder does not change its choice. A linear model using only trace increments recovers the correct choice…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22285v1 Announce Type: new Abstract: Adapting language models to new domains via continual pre-training raises a basic evaluation problem: if the training corpus overlaps with what the model already knows, performance gains cannot be cleanly attributed to new learning rather than pre-existing knowledge. This matters most for knowledge-intensive, task-light (KHTL) robot deployments - pharmaceutical dispensing, hazardous-material handling, facility-specific protocols, where the physical task is simple but the governing rules are proprietary and safety-critical, and where extensive live testing is costly or unsafe. We introduce ORDER (Ontology-driven Decision-making for Embodied Reasoning), a benchmark built on a fictitious world: a 342,069-token synthe…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22278v1 Announce Type: new Abstract: Quadruped manipulators enable mobile grasping in complex environments, yet their whole-body control policies remain vulnerable to unreliable onboard visual perception. Existing methods are typically developed under relatively reliable observations and have not systematically examined how occlusion, segmentation-mask dropout, depth noise, and target-localization jitter affect grasp reasoning and target tracking. To address this gap, we introduce DeViGrasp-Bench, a benchmark for mobile grasping under degraded vision that incorporates controlled visual degradations, seen and unseen objects, multiple difficulty levels, and complex terrains, and evaluates task success, execution efficiency, and action smoothness. We fu…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22276v1 Announce Type: new Abstract: The Dynamic Window Approach (DWA) is widely used for local navigation, but its performance depends strongly on parameters that are typically fixed before navigation. In particular, the appropriate prediction horizon can vary with local free space: longer horizons support efficient motion in open areas, whereas shorter horizons help preserve feasible motions in narrow or cluttered regions. This paper proposes D3DWA, an adaptive DWA framework based on a Dueling Double Deep Q-Network (D3QN), which jointly selects the DWA evaluation weights and prediction horizon from a continuous navigation state at every control step while retaining DWA's trajectory generation and collision checking. In eight simulated environments,…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22295v1 Announce Type: new Abstract: Event pipelines often discretize asynchronous timestamps before learning. This step looks harmless, but its stability depends directly on temporal resolution. We study this dependence at the representation level. We first show that hard temporal binning is discontinuous: an arbitrarily small timestamp shift near a boundary can move unit event mass between bins. We then define a class of nonnegative, mass-preserving, resolution-faithful continuous encoders and prove that every encoder in this class has global L1 sensitivity at least 2/Delta, where Delta denotes bin width. Linear two-bin interpolation attains this limit. Local support and first-moment preservation also make it unique. Experiments on SHD, N-MNIST, an…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22284v1 Announce Type: new Abstract: Talking-face (TF) deepfakes are detected unevenly by rPPG-based methods across generators. We study two lightweight visual-only cues, rPPG-derived waveforms extracted by RhythmFormer and lip-region discrete cosine transform (DCT) coefficients, on the seven TF methods of Celeb-DF++ under a subject-independent protocol. In-domain, lip-region DCT matches or exceeds the rPPG-derived 1D ResNet on every method except SadTalker, and Concat fusion reaches AUC 0.891 against 0.824 and 0.827 for the unimodal baselines. Under leave-one-generator-out evaluation the cues split: each transfers clearly better to three held-out methods, and IP-LAP is near chance for both. Concat averages 0.798 but falls below rPPG alone where DCT…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22267v1 Announce Type: new Abstract: Camera motion often reflects directorial intent and requires professional equipment, making it a high value form of intellectual property. However, generative video models can imitate such high value camera motions with simple prompts, while existing similarity detection methods mainly operate on visual content and fail to capture deeper motion similarity. This is mainly because their training data entangles camera motion with visual content. Moreover, traditional optical flow is insufficient to represent complex camera motions. We therefore build the first benchmark for camera motion analysis, including a motion dataset with \textbf{11} motion styles and evaluation protocols. Furthermore, we propose a motion repr…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22096v1 Announce Type: new Abstract: Climate campaigns are often evaluated through attention and mobilization, but less is known about the well-being language that accompanies them. Whether campaign periods alter positive affect and hope, and whether happiness aligns with action language, remains unresolved. We analysed 364,118 public Twitter/X posts from Earth Day, Earth Hour, Global Climate Action Day and World Environment Day in 19 occurrence-years, using 30-day pre-event, event and post-event windows. A versioned weighted lexical model estimated happiness, future-oriented hope, collective capability, distress and action language. Event-period happiness prevalence was 9.02 percentage points higher than the pre-event baseline , whereas paired occur…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22122v1 Announce Type: new Abstract: Cross-device hardware evaluation often assumes that if architecture rankings transfer across devices, a proxy device can support target-side model selection. We stress-test this assumption for joint latency-energy feasibility across two public architecture families. On NAS-Bench-201, cross-device rank correlations are moderate, while target-comparable feasible-set overlap remains incomplete. A faithful AdaProxy diagnostic substantially improves latency ranking, showing that the observed boundary failures are not simply due to weak adaptation. Exact finite-sample split-conformal analysis also exposes an evidence bottleneck: a finite one-sided 90% threshold requires at least nine calibration observations. We then re…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22113v1 Announce Type: new Abstract: Persistent low retention and completion rates in medications for opioid use disorder (MOUD) have driven the use of machine learning (ML) models to predict retention and identify patients at risk of premature discontinuation. However, the fairness of these models across patient populations remains largely unexplored, raising concerns about their application in treatment decision support. This study systematically assesses algorithmic fairness in ML models for predicting MOUD retention and premature discontinuation and investigates the effectiveness of bias mitigation techniques. Using the cross-sectional Treatment Episode Data Set-Discharges (TEDS-D), which includes treatment episodes for individuals in the U.S. di…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Citizens Advice says providers of services such as energy, banking, phone and internet should guarantee the ‘right to talk to a human’ Stop relying on AI chatbots for customer care and guarantee the “right to talk to a human”, Citizens Advice has urged essential service providers, as research found they wasted time, caused stress and delayed problem solving for more than half of users. The spread of the AI-powered systems to provide help about the provision of vital services such as energy, banking, phones and internet, is making it harder for millions of already digitally excluded people to handle snags, according to the frontline charity that last year provided more than 2.7 million people with one-on-one help in England and Wales. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Donald Trump and Xi Jinping will meet each other next week, but it’s unlikely they’ll agree to AI regulation Hello, and welcome to TechScape. I’m your host, Blake Montgomery, US tech editor at the Guardian, writing to you while reading and disagreeing with Jenny Odell’s 2019 book How to Do Nothing. I’m pro-screen time. Today in tech, we’re discussing the upcoming meeting between Donald Trump and Xi Jinping, in which AI will likely feature prominently. Continue reading...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22108v1 Announce Type: new Abstract: Learning-enabled perception is important in many autonomous systems. Unlike traditional sensors, the boundary where ML perception does or does not work is poorly characterized. Incorrect perception can lead to unsafe or overtly conservative downstream control actions. In this paper, we propose a two-step strategy for correcting ML-based state estimation. First, an offline computation is used to characterize the uncertainties resulting from the ML module's state estimation, using preimages of perception contracts. Second, at runtime, a risk heuristic is used to choose particular states from the uncertain estimates to drive the control decisions. We perform extensive simulation-based evaluation of this runtime perce…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：This article is brought to you by CoolIT. Beyond 250 kW a server rack can no longer be cooled by a hybrid approach of liquid and air. At this density a 70/30 liquid-air split leaves 75 kW of air load. The air cooling system needed to move it brings cost and complexity few operators will accept. The answer is near-total heat capture. Liquid takes effectively all the heat, air falls below 1 percent of the load, allowing the server to run fanless. CoolIT builds these loops today from modular coldplate blocks proven across six generations of fanless designs. Processor thermal design power (TDP) keeps climbing generation over generation. This rising heat load is now cascading into the memory, networking, storage, and power components that once ran comfortably on air…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22332v1 Announce Type: new Abstract: Generalizable robot manipulation requires predicting how a scene will evolve, identifying where interactions are feasible, and determining how to act. Action-labeled robot videos directly supervise control but are costly and limited in diversity, whereas egocentric human videos capture diverse interactions but lack robot actions and differ in embodiment and appearance. We introduce AffordanceWAM, an affordance-aware generative World Action Model that represents object-centric spatiotemporal affordance through Scalar Affordance and Affordance Heatmap, within the generated future World. This representation grounds visual prediction in task-relevant objects and interaction regions for action generation, and provides…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22289v1 Announce Type: new Abstract: Large-scale egocentric and whole-body human demonstrations are becoming a primary source of data for embodied intelligence. They record what people perceive and do, but rarely the external specification that gave an action its purpose. In construction that omission is consequential: skilled work is directed at project-specific configurations defined in a design model - configurations not yet present in the environment being observed. A mason's transferable competence is not the geometry of one wall but the ability to realise a new geometry from a specification. We introduce the specification-conditioned demonstration: a synchronised record of the physical state a demonstrator perceives, the intended state supplied…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2609.22274v1 Announce Type: new Abstract: Recent advances in humanoid robotics have produced diverse skills through reinforcement learning, motion imitation, and generative modeling. Yet these capabilities remain siloed because they are built around incompatible representations, interfaces, and controllers. We present CHOREO, a framework for training-free composition of heterogeneous humanoid skills. Our key observation is that, regardless of how a skill is learned, it can ultimately be expressed as an executable motion trajectory. Based on this observation, CHOREO converts each capability into SkillMotion, a unified representation that combines motion states, contacts, semantics, and boundary conditions. Skills are composed through direct continuation, c…