創業融資

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：The California Department of Health Care Services (DHCS) built the Behavioral Health...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Company, which grown rapidly in recent years, has been under scrutiny as privacy concerns grow Flock Safety plans to let go of about 18% of its employees, people with direct ⁠knowledge of the plans said on Thursday, as the maker of AI-powered surveillance cameras and license-plate readers faces mounting opposition to its products from communities and ⁠lawmakers. The people said the job ⁠cuts came after ​a voluntary buyout program and were expected to affect roughly 270 employees at Flock, a surveillance technology startup that has seen rapid growth in recent ⁠years. Employees will leave the company at the end of the month. 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："Staggering." "Overwhelming." "Unprecedented." "Surreal." "Pure insanity." Those were among the descriptions more than three dozen mathematicians reached for in conversations with The Verge as they tried to make sense of the flood of mathematical results OpenAI abruptly dropped on the field this week. Amid the awe, excitement, and uncertainty over the sheer scale of the deluge was a deep-seated anxiety over what it all means - and what comes next. For all their different reactions, researchers agreed that simply understanding what OpenAI had released could take years, let alone figuring out where the mathematicians themselves fit in the fi … 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：Cloudflare is buying the startup founded by Node.js creator Ryan Dahl, a longtime competitor that recently built its own open-source The post Cloudflare acquires Node.js creator’s startup that copied its serverless playbook 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：The original first place video used AI in post-processing. | Image: Dr. Ning Xu Nikon says the video that originally won first place in its Small World in Motion contest "did not comply with the competition rules regarding generative AI." BBC reports that the original first place video from Dr. Ning Xu claimed to show "tiny, hair-like structures called cilia moving in the airway of a child with the respiratory condition PCD." Nikon said last week that it was reviewing the video following skepticism online about its authenticity. In a comment on LinkedIn, Dr. Xu admitted to using AI for the video: "An unsupervised neural-network method was subsequently used for AI-assisted post-processing to distinguish and visualize f … 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：A monthly recap of the latest Amazon Bedrock, Amazon Bedrock AgentCore, and Strands updates from September 2026: broader model choice, faster serverless agents with built-in evaluation, and automated knowledge base syncing with native enterprise connectors.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：When every tier of the defense supply chain can share governed data with the people who need it...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Instinct’s agent is always just a text away. Before there were cute little guys, there was Instinct. In August, the startup got its AI agent to market with an unusual playbook: invite-only, no marketing, and barely so much as a website. And yet, Instinct quickly became the buzziest thing in AI, garnering praise for its straightforward, text message-based interface and its ability to handle chores like booking DMV appointments and sending follow-up emails. Then Muse arrived, followed not long after by Dots. The same products, more or less, from two far more powerful companies. With Big Tech players suddenly in the mix, it was looking dubious that the startup's buzzy launch could keep … 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：Questions raised over AI growth as ChatGPT maker forecasts this year’s revenue at $50bn, way below the $70bn signalled before OpenAI has revealed that it is making about $20bn less in projected revenue than it had recently indicated to investors, raising questions about the break-neck growth rate in demand for AI. The ChatGPT-maker company has told investors that its revenues for this year would reach $50bn (£37bn), a projection based on sales up to the end of September. 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：Former Labour deputy leader says tech firm would work with a Reform UK government on immigration crackdowns Tom Watson, the former Labour deputy leader who recently joined Palantir, has warned against “mob rule” when it comes to awarding public contracts. Lord Watson, now a senior vice-president at the US tech corporation, warned UK ministers could get themselves “in a lot of trouble” as he was questioned about concerns raised about the government working with his new employer. 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:2610.10637v1 Announce Type: new Abstract: Hair stroking is common in daily grooming and personal care, and is also widely used in hair-product evaluation, motivating robots with similar physical interaction capabilities. Existing robotic hair-care and surface-following methods mainly rely on trajectory planning, compliance, force regulation, or tactile-conditioned policies, but deformable hair can remain in contact while gradually drifting across the end-effector, making local interaction difficult to regulate. We propose TacHair, a tactile contact-distribution guided online correction framework that represents high-resolution tactile observations as a spatial hair-contact distribution. A visuotactile imitation policy generates the nominal stroking motion…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.10703v1 Announce Type: new Abstract: Eyeglass reflection removal is important across smartphone imaging, video conferencing, and other face-centric visual applications. The task is challenging because reflections range from mild photometric contamination to severe ocular occlusion, requiring selective correction and plausible reconstruction without altering identity or natural appearance. Existing datasets cover limited reflection conditions, constraining generalization to complex real-world scenes and systematic evaluation. We introduce \textbf{OcuBench}, a multi-source benchmark comprising 10,280 controllable synthetic pairs, 732 real-input pseudo-pairs, and 458 independent real-world test images, supporting both paired evaluation and assessment be…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.10827v1 Announce Type: new Abstract: Corrective feedback is among the best-evidenced drivers of second-language acquisition, yet corrections delivered during lessons rarely accumulate into an actionable view of grammar mastery. Prompted frontier models can provide such a view from learner--tutor lesson transcripts, but they are costly at scale. We close this gap by fine-tuning Qwen3.5 small language models (SLMs) on filtered and rebalanced teacher-generated supervision, then deploying an efficient 0.8B model in an end-to-end grammar mastery tracker for all English learners on our platform. Internalizing the annotation contract into adapter weights enables pairing the 0.8B model with a compact matched prompt rather than verbose instructions. On two hu…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.10594v1 Announce Type: new Abstract: Activation probes that monitor deployed language models are trained on synthetic conversations, and how many a probe needs is open. We trace learning curves over 10-590 synthetic samples for three monitoring concepts, high-stakes situations, replies harmful to a person, and replies that do not follow the user's instruction, on fourteen held-out evaluation distributions and four probe models, varying the generator LLM and the prompt's detail. The need is set by what is monitored: probes for high-stakes and harmful are within a few hundredths of their plateau from 80 samples on Gemma-3-27B-IT, instruction probes need several times as many, and the ordering holds on three smaller probe models and on real samples (fro…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.10590v1 Announce Type: new Abstract: Tool-using agents repeatedly carry observations whose useful content can be much smaller than their original payload. We study agent-controlled forgetting: the acting model selects previously observed tool results, replaces each with a short note at its original position, and retains the exact original in a recoverable archive. A Python harness exposes batch archival and explicit recovery without task-specific model training, while protecting user instructions and assistant messages from these operations. In an exploratory OpenTelemetry debugging case followed by an unrelated implementation task, the method ended with 231,951 provider-reported prompt tokens versus 912,492 under retained history, used 50% fewer cum…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.10541v1 Announce Type: new Abstract: Knowledge Graph (KG) quality depends not only on downstream graph validation, but also on the quality of tabular metadata used before integration. In metadata-only Semantic Table Interpretation (STI), where cell values are unavailable, noisy, or unsuitable, column headers become a critical source of semantic evidence for traceable KG preparation. We present an explainable, header-centric framework for metadata-only Column Type Annotation (CTA) and Data Quality Assessment (DQA). The framework maps headers to 39 interpretable FinalFormat types using curated lexical resources and preserves token-level traceability through SourceKeywords. Each assigned type activates validation rules based on a taxonomy of Data Qualit…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Discover how Snyk transformed an internal support agent into Snyk Assist, a customer-facing AI feature powered by LangChain, LangGraph, and LangSmith.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：The $11-a-share offer would have been the biggest debut on the stock market since the telecommunication giant in 1997 Get our breaking news email, free app or daily news podcast Firmus Technologies has scrapped what was set to be Australia’s biggest company listing in decades after investor demand for its much-hyped AI datacentre business failed to materialise. A Firmus spokesperson said the board decided that proceeding with the offer was no longer in the best interests of the company and its shareholders. 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：The California State Athletic Commission sent a cease-and-desist letter to a startup that hosted a match between a human and a robot last month, as reported by The New York Times. The fight, which took place on September 18th, pitted a human, Frankie LaPenna, against a humanoid robot owned by a tech startup, Rek, that was being piloted by a human using what the NYT described as a "remote virtual-reality system." Rek says it is the "the humanoid robot fighting league" on its website, and the robot appears to be one from EngineAI but with a Terminator-like head swapped on top. You can watch a replay of the fight on YouTube: The CEO of Rek, … 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：My Decoder guest today is Hayden Field, The Verge’s senior AI reporter, and we’re discussing the new wave of consumer-friendly AI agents. If you’ve been paying attention to this space, you know AI enthusiasts have been using agents for a minute now — homebrew OpenClaw setups led to a surge in Mac Mini sales earlier this year. But the launch of Meta’s Muse, OpenAI’s Dots, and xAI’s Grok bot has brought easy to use agents to millions. Muse and Dots have had the highest-profile product launches, and they’re fascinating to pit against each other. Both Meta and OpenAI have decided to pitch these agents to mainstream users and businesses in the form of cute, animated mascots. Verge subscribers, don’t forget you get exclusive access to ad-free Decoder wherever you get…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：OpenAI, CloudFlare, and Amazon have all copied this tiny startup.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：With GPT-5.6, GPT-6 Astra, and GPT‑Image‑2.5, Pollo AI helps creators turn bold ideas into detailed images and cinematic video ads.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Sparse neural retrieval finally started getting more and more attention (SPARSEUP, MILCO, sparse encoders in Sentence Transformers…). Well, amazing, we like attention! Our miniCOIL v1 sparse neural retriever article ended with a promise: to keep working on this sparse neural retriever in-depth, improving the model’s quality, and in-width, “extending it to other dense encoders and to languages beyond English.” Here is the next “in-width” step of the miniCOIL saga: to try and cross the language barrier, that is, make the model suitable for multilingual retrieval. It’s a fun challenge, as sparse retrievers are hard to adapt to this scenario: exact matching in cross-lingual text search usually means translation.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08970v1 Announce Type: new Abstract: Humanoid loco-manipulation of large, heavy objects demands forceful interaction across the entire body. However, such payloads shift a humanoid's center of mass and impose sustained loads across the upper body, challenging balance and command tracking. We present HULK, a whole-body control framework for forceful loco-manipulation. Using model predictive control (MPC) to guide reinforcement learning with predictions of the loaded dynamics, we train two teachers: one tracks arm motions under wrist forces, and the other locomotes while holding large objects against the body. A capture-point control barrier function augments the wrist-force teacher during training to improve balance under load. We distill both teacher…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08800v1 Announce Type: new Abstract: Models based on graphs have emerged in robotics as a powerful foundation for internal world representations, where factor and scene graphs are among the most prominent model types found in the related literature and in successful robotic solutions. Initially, many of these models were assuming static environments as a simplification. Herein, factor graphs mainly provide uncertainty-aware geometric estimations while scene graphs enable a structured semantic abstraction. However, real-world robotic environments are often dynamic, posing severe challenges for purely static world representations. Therefore, this review presents a comprehensive view on how dynamic aspects of real-world environments can be addressed in…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08983v1 Announce Type: new Abstract: Generative inpainting of brain MRI volumes is essential for synthesizing healthy tissue in pathological regions, improving the accuracy and reliability of automated downstream brain analysis applications such as image registration, brain extraction, and segmentation. However, standard 3D approaches are computationally prohibitive, while efficient 2D slice-wise methods suffer from severe inter-slice discontinuities. Furthermore, traditional models rely on conditional training, requiring task-specific learning of masked inputs. We propose a zero-shot brain MRI inpainting framework utilizing 2.5D unconditional flow priors to capture spatial context along the superior-inferior axis without the overhead of full 3D conv…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08944v1 Announce Type: new Abstract: Scanner variation changes how pathology foundation models represent the same tissue. We introduce SlideRuler, which uses regions within a slide as internal controls to estimate and correct acquisition-induced shifts in other regions. A transfer map learned from paired rescans enables calibration from a single scan at inference while keeping the foundation model fixed. Across two encoders and five SCORPION scanners, learned transfer reduces mean target-to-source embedding distance by 16.3-38.5% relative to raw embeddings. Comparisons with unrelated same-scanner controls reveal a positive same-slide contribution across all four evaluation settings, including scanner holdout. A source-anchored variant reduces source-…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08936v1 Announce Type: new Abstract: Robustness of image classification has several benchmarks, but their video counterparts are absent. In video classification temporal dimension introduces additional degrees of freedom for adversarial attacks, defenses, and preprocessing. Temporal sampling, perturbation budgets, and metric aggregation also interact in ways with no direct analogue in the image setting. Therefore, robustness for video classifiers is studied across scattered, incompatible implementations, making reported numbers hard to reproduce and analyze. We introduce VCR-Bench, a modular open-source benchmark framework that standardizes video loading, wrappers for classifiers, adversarial attacks and defenses, perceptual metrics, configuration pr…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08825v1 Announce Type: new Abstract: Autonomous driving has made remarkable progress, with recent AI advances enabling commercial deployments that are reshaping urban mobility. Yet the field remains far from its universal social promise: autonomous systems that can operate robustly anywhere, anytime, for anyone. We posit that this gap is not merely a modeling problem, but a problem of the prevailing data paradigm. Current research relies heavily on a few benchmark datasets with limited spatial and scenario coverage, even though the community has collectively produced over 600 autonomous driving datasets across nearly 50 countries. However, this abundance has not translated into broad research impact: most datasets remain significantly underused due t…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08813v1 Announce Type: new Abstract: X-ray image-based Radiology Report Generation (RRG) constitutes a critical research direction within medical artificial intelligence, with great potential to alleviate clinicians' diagnostic workload and shorten patient waiting periods. Despite substantial advances over recent years, the field faces evident bottlenecks stemming from insufficient standardized benchmarks and inadequate domain adaptation of generic large models. Notably, the newly released CheXpert Plus dataset is provided without accompanying baseline implementations and evaluation results, which impedes standardized training, quantitative evaluation and fair comparison among follow-up algorithms. To mitigate this limitation, we establish a comprehe…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08828v1 Announce Type: new Abstract: Small-data adaptation can improve speech detection while degrading speaker attribution. We study this discrepancy in a released streaming diarizer adapted on 7.5 h of two-party conversation and evaluated across six corpora. Adaptation substantially improves in-domain diarization performance and transfers to an independent corpus. However, this improvement is not consistent across evaluation scenarios as the additional confusion is mainly associated with impaired temporal identity consistency rather than speaker-count errors. A local-remapping diagnostic reveals different patterns of identity degradation across corpora, indicating that adaptation may alter how streaming models maintain speaker assignments over time…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08794v1 Announce Type: new Abstract: Large language models rely on subword tokenizers whose quality varies across languages, yet no standardized multi-metric framework exists for broad comparative evaluation. We introduce Tokka-Bench, an open-source framework that evaluates tokenizers on five complementary metrics -- bytes per token, unique token coverage, subword fertility, word-split rate, and vocabulary composition -- across 100 natural languages (30+ scripts) and 20 programming languages, using language-aware segmentation adapted to each writing system. Comparing seven BPE tokenizers (GPT-2, GPT-4, gpt-oss, Llama 3.1, Gemma 3, Qwen3, and Kimi K2) within individual languages, we find that vocabulary allocation strategy matters more than raw vocabu…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.08810v1 Announce Type: new Abstract: Fine-tuned geospatial foundation models (GeoFMs) pretrained on large satellite archives have been shown to improve crop classification accuracy and geographic transferability. However, their operational performance beyond the training distribution remains poorly characterized. We evaluated the out-of-distribution performance of a widely adopted GeoFM [Prithvi-EO-2.0] across 37 events in 12 countries on three continents and validated against regional reference products. Results indicated that the mean overall accuracy (OA) declined from 0.65 in the United States to 0.40 in Europe. Beyond accuracy metrics, we assessed five key aspects of model performance: whether model confidence indicates signal failure, sensitivi…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Sources say Firmus is slashing its price and may even shelve initial public offering altogether Get our breaking news email, free app or daily news podcast The momentum behind Firmus Technologies’ high-flying valuation is showing severe cracks just weeks out from its anticipated ASX debut. Multiple sources briefed on the matter told Guardian Australia the AI datacentre company is slashing its valuation to entice sceptical investors – or may even shelve its initial public offering altogether. 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：Nvidia-backed open source startup makes splashy entrance. AI hits U.S. K-8 education.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Deep Agents now lets you bind tools to skills, pin skills at runtime, and reload skills mid-thread, so agents with expansive skill repositories stay context-efficient and effective.

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Databricks Apps lets developers build and deploy data and AI applications directly on the Databricks platform...

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：Google DeepMind, Meta, and AI drug discovery startup Isomorphic Labs are jointly investing $300 million into Biohub, the nonprofit biomedical research organization founded by Mark Zuckerberg and his wife, Priscilla Chan, as reported by Reuters. The funding is part of a $1.8 billion initiative to build AI datasets that could allow researchers to "ask, predict, and answer biological questions digitally," helping find new ways to prevent and manage diseases. Founded in 2016, Biohub aims to combat diseases through the creation of a "virtual cell" that researchers can use to carry out simulations. To support this effort, the US Department of Ene … 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：DoorDash, the leading food delivery app, processed 970 million orders in its second quarter this year and generated $4.5 billion in revenue. A 10-person startup called Bites is a blip in comparison: It has just around 300 restaurants signed up in the Bay Area, where it's operating as a pre-seed startup. But this summer, Bites caught DoorDash's attention. In August, a slew of restaurants in the Bay Area received a strange email from DoorDash, warning businesses that they may be listed without their consent on Bites. The form letter, copies of which were seen by The Verge, said that DoorDash had heard from "several partners" that were adde … 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：Common Sense Media, a nonprofit that offers reviews of apps, services, and entertainment with a focus on youth safety, today said that OpenAI's ChatGPT for Teens is an "unacceptable risk." ChatGPT for Teens, introduced in August, has guardrails for teens and is designed to help students learn, but Common Sense Media says that the teen protections offered by the feature "fall short of what the company promised." Common Sense Media's assessment found that ChatGPT "doesn't send alerts to parents when it should," doesn't offer "the right help in crisis situations," and "still does kids' homework," according to a press release. "ChatGPT for Tee … 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：Meta has open-sourced Rebalancer, the C++ and Python library it has used for over 9 years to place shards, servers and traffic. It handles about 40 million assignment problems a day, using local search or MIP solvers like Gurobi, FICO Xpress and HiGHS. It is pip-installable under Apache 2.0. The post Meta AI Open-Sources Rebalancer: A C++ Assignment Solver That Runs About 40 Million Placement Problems a Day 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:2610.06882v1 Announce Type: new Abstract: We study assignment quality in 3D heterogeneous multi-agent reach-avoid games and identify a recurring failure mode of cardinality-only matching in geometrically structured scenarios, which we term \emph{Geometric Sprawl}. In these cases, multiple maximum-cardinality assignments are available, but some induce spatially incoherent pairings and inefficient pursuit trajectories. Building on the evasion-space framework of Yan et al.~\cite{yan2022}, we introduce a cardinality-first weighted sequential matching method in which the Hamilton--Jacobi--Isaacs interception value $z_I(s,j)$ is used as a secondary assignment weight. Each sequential stage is solved with a min-cost max-flow backend, while the unweighted baseline…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.06862v1 Announce Type: new Abstract: **Context:** Deep Neural Networks (DNNs) increasingly control Cyber-Physical Systems (CPSs), yet small input perturbations can cause unsafe system-level behavior. Existing approaches often optimize perturbations for individual images and evaluate them only in simulation, limiting their generalizability and practical validity. **Objectives:** This work aims to generate robustness tests that remain effective across operational observations and to evaluate whether the resulting failures transfer from simulation to a physical robot. **Methods:** We propose an explainability-guided multi-objective evolutionary approach that generates sparse perturbations over representative images selected through visual and behavioral…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.06973v1 Announce Type: new Abstract: Melt-pool monitoring is central to qualifying metal additive manufacturing (AM), yet no public event-camera benchmark exists for this domain. Event cameras report per-pixel brightness changes with microsecond timing instead of reading full frames, giving the temporal resolution AM transients demand at a fraction of the data rate. We present SynAM-E (Synthetic AM Events), the first public multi-source simulated event-camera benchmark for metal-AM melt-pool monitoring: 85 physics-calibrated event shards from 15 sources across 8 institutions, with public baselines and fixed cross-machine evaluation splits. On a single-machine case study, event-spatial monitoring matches dense-frame accuracy (0.874 versus 0.863 macro-…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.06972v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in visual understanding and multimodal reasoning, yet they remain fundamentally limited in Human Action Feedback Generation. Existing methods infer coaching feedback directly from visual observations, producing generic advice, limited interpretability, and physically implausible hallucinations. In contrast, expert human coaches diagnose performance through explicit biomechanical reasoning over joint kinematics, posture, and body dynamics. We introduce BoT-Feedback, a framework that grounds MLLM reasoning in structured biomechanical evidence. Our key contribution is Biomechanics of Thought (BoT), a four-stage reasoning framework that…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.06960v1 Announce Type: new Abstract: We present DistScene, a framework for single-image compositional 3D scene generation by jointly modeling the environment and individual objects. Unlike existing methods that represent scenes primarily as collections of objects, we model the environment as an explicit scene component to provide geometric context for object placement. Specifically, we introduce Scene-Frame Generation, which jointly generates separate environment and object components in a shared coordinate frame, allowing their geometry and relative placement to be learned together. Then we introduce Object-Centric Refinement to refine each object in a local frame with scene context. Finally, we develop Object-to-Scene Distillation to transfer pretr…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.07019v1 Announce Type: new Abstract: Fiorillo v0.5 is an open model that answers typed questions with a probability for each answer. Its main specialist reads a randomized trial's article, cut to 6,144 tokens, and answers whether an intervention significantly increased, significantly decreased or did not significantly change an outcome against a comparator (Evidence Inference 2.0, EI). It is Qwen3-4B-Base with low-rank adapters and a decision head, fine-tuned for EI only on the 1,431 of 2,657 training articles whose own license allows reuse. Four criteria registered on the Open Science Framework before this version's test predictions decided its release, the second bar judged on EI's test split, whose labels are public. On that split (1,218 prompts i…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.06956v1 Announce Type: new Abstract: Large speech language models have demonstrated strong capabilities in unified cross-modal understanding and generation, yet paralinguistic cues, especially emotion, remain difficult to preserve. Existing systems typically rely on entangled acoustic representations, which allow the underlying language model to depend excessively on recovered lexical content instead of grounding its behavior in acoustic-prosodic evidence. We address this limitation with EMODE, an emotion-aware speech language model built around \textbf{Dynamic Para-Semantic Experts (DPSE)}. DPSE decomposes continuous speech features into semantic and paralinguistic pathways, routes them dynamically, and fuses them before integration into the languag…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.06902v1 Announce Type: new Abstract: Retrieval-augmented generation grounds language models in external context, but for long documents flat top-$k$ retrieval can cluster on a single region and miss complementary evidence. RAPTOR-style summary trees address this by recursively clustering chunks and using a language model to summarize each cluster at indexing time, then ranking summary nodes alongside raw chunks at query time. We show the main benefit of summary trees in long-document QA can come from navigation rather than the generated summary content. We introduce NavTree, a leaves-only retriever that builds a deterministic balanced segment tree over chunks (zero language-model calls at indexing) and uses the tree purely as a navigation scaffold: a…

AI 服務暫時不可用，以下為來源摘要，待恢復後補全翻譯：arXiv:2610.06890v1 Announce Type: new Abstract: We present an industry experience report on three years of operating an event-driven cloud infrastructure for continuous machine learning training in automotive manufacturing. Our system orchestrates GPU-accelerated training of product-specialized model pairs, a physics prediction model and a reinforcement-learning control policy, across multiple plants, coordinating long-running GPU workloads triggered by manufacturing events. The architecture combines Amazon ECS with EC2 GPU capacity providers, SQS-based messaging with dead-letter queues, and an admission-controlled Lambda dispatcher that enforces cluster concurrency limits. A Conductor orchestrator on ECS Fargate initiates dependency-aware retraining chains on…