AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchRevealing what benchmark scores hide: Jointly measuring the breadth and reliability of LLM capabilities using accessibility profiles
arXiv:2610.24957v1 Announce Type: new Abstract: Language model benchmarks assign each question a single outcome from one fixed wording, even though the same underlying request can be expressed in many equivalent ways. We…
- arXiv cs.AIresearchTowards Human-AI Complementarity for AI Safety
arXiv:2610.24969v2 Announce Type: replace Abstract: Safely overseeing AI is increasingly challenging for the humans doing it, as they must evaluate longer and more complex outputs with every new model generation. To miti…
- arXiv cs.AIresearchCorrectRTL: Boost RTL Generation Quality Through Automated Self-Correction with Instruction Decomposition
arXiv:2610.24977v1 Announce Type: new Abstract: Large language models (LLMs) have shown strong promise in specification-to-RTL generation. Among the explorations, the self-correction paradigm has significantly advanced L…
- arXiv cs.AIresearchVLA-Rolling: Fewer Denoising Steps, Better Control via Adaptive Action Reuse
arXiv:2610.25043v3 Announce Type: replace Abstract: Flow-matching vision-language-action (VLA) models generate an action chunk by iteratively denoising Gaussian noise, but typically execute only a short prefix before rep…
- arXiv cs.AIresearchSteadyThought: Mitigating LLM Under-Thinking via Thought-Level Preference Optimization
arXiv:2610.25045v1 Announce Type: new Abstract: Flexible switching between reasoning trajectories (i.e., thought switching) has significantly enhanced the reasoning capabilities of Large Reasoning Models (LRMs). However,…
- arXiv cs.AIresearchSpectral Adaptive Graph Transformer for Efficient and Scalable Traffic Forecasting
arXiv:2610.25140v1 Announce Type: new Abstract: Traffic forecasting models based on spectral graph convolution offer an interpretable account of how information is mixed between sensors, but they rely on an eigendecompos…
- arXiv cs.AIresearchDialogue Generation for Conversational Recommendation Using User Search Interactions and Knowledge Graphs
arXiv:2610.25182v1 Announce Type: new Abstract: Large-scale conversational recommendation datasets typically require manual dialogue construction or annotation. We propose SearCl, which automatically generates conversati…
- arXiv cs.AIresearchAligning Language Models with Indigenous Community Cosmovision through Constitutional AI
arXiv:2610.25298v1 Announce Type: new Abstract: Large language models (LLMs) are predominantly trained on corpora that reflect Western, anglophone worldviews. This asymmetry is not neutral: when a model answers questions…
- arXiv cs.AIresearchWhen Form Changes but Logic Doesn't: Building Logic-invariant LLMs through Logical Structure
arXiv:2610.25402v2 Announce Type: replace Abstract: Large language models (LLMs) have achieved strong performance across diverse reasoning tasks, yet it remains unclear whether this performance reflects an understanding…
- arXiv cs.AIresearchStepwise Reasoning Verification: Don't Generate, Classify!
arXiv:2610.25441v1 Announce Type: new Abstract: Verifying the reasoning of large language models is increasingly important as they are applied to complex mathematical and scientific problems. Existing work focuses on lar…
- arXiv cs.AIresearchOn Reasoning Correctness Directions
arXiv:2610.25464v1 Announce Type: new Abstract: We present experimental evidence suggesting that information about whether a model's reasoning is on the right track can be linearly decodable without defining a causally u…
- arXiv cs.AIresearchA failure of Imagination: VLMs fail to use visual representations for text, but self-distillation can help
arXiv:2610.25470v1 Announce Type: new Abstract: Vision-language (VL) models are typically built by fine-tuning a language model to attend to visual features in service of a Visual Question-Answering objective. We ask whe…
- arXiv cs.AIresearchP(Yes): 15$\times$ Improvements via Test-Time Scaling of Decision Models
arXiv:2610.25474v1 Announce Type: new Abstract: Decision models answer typed questions, such as multiple-choice or yes/no questions, with a probability for each option. We add test-time search to frozen decision models,…
- arXiv cs.AIresearchFrom Proposal to Execution: Modeling Interaction Semantics in Shared Autonomy
arXiv:2610.25496v2 Announce Type: replace Abstract: In many human-AI systems, candidate actions are not executed directly but are first evaluated by another decision-making entity under a potentially different objective…
- arXiv cs.AIresearchAssessing the AI Readiness of Construction Datasets: A Multidimensional Operational Framework
arXiv:2610.25499v1 Announce Type: new Abstract: The rapid growth of artificial intelligence (AI) applications in construction has increased the demand for datasets suitable for reliable model development, evaluation, and…
- arXiv cs.AIresearchFormal Raising: Embedding Verifiable Ethical Properties into the Development of AI
arXiv:2610.25501v1 Announce Type: new Abstract: Alan Turing proposed constructing a ``child machine'' and educating it rather than directly programming an adult-like intelligent machine. Recent work on model raising revi…
- arXiv cs.AIresearchWhen Empathy Misses the Goal: A Benchmark for Goal Displacement in LLM Advice
arXiv:2610.25522v1 Announce Type: new Abstract: Conversational large language models can respond with warmth and reassurance while failing to advance the user's stated goal. We conceptualize this failure as $\textbf{goal…
- arXiv cs.AIresearchFrom Clinical Risk Prediction to Operational Decisions: An Uncertainty-Aware Framework for ICU Staffing
arXiv:2610.25530v1 Announce Type: new Abstract: Advances in electronic health records (EHR) and large-scale critical care databases have enabled the development of machine learning models for the early detection of patie…
- arXiv cs.AIresearchNarrow Decision Models: A System-1 Architecture for Calibrated, Typed Decisions
arXiv:2610.25553v1 Announce Type: new Abstract: System One models, introduced with TypeSafe's Jev, answer typed questions about a state with calibrated probabilities instead of generated text: choosing among options desc…
- arXiv cs.AIresearchThe Forward/Inverse Duality of Controlled Emergence:A Computational Design Program from Explanation to Engineering
arXiv:2610.25630v1 Announce Type: new Abstract: The macroscopic capabilities of intelligent systems arise from the local interactions of many simple units. When such systems are deployed in the real world, their behaviou…
- arXiv cs.AIresearchThink, Compress, Answer: Reinforcement Learning for Multimodal Grounding via Reasoning Compression
arXiv:2610.25640v1 Announce Type: new Abstract: Extended reasoning has substantially improved multimodal large language models (MLLMs) on tasks requiring compositional and relational inference. However, multimodal ground…
- arXiv cs.AIresearchSpherical Fixed-Point Reasoners: Adaptive Loop Halting on the Hypersphere
arXiv:2610.25646v1 Announce Type: new Abstract: Looped (weight-tied) transformers enable adaptive computation by allocating more iterations to harder inputs. Fixed-Point Reasoners (FPRM) achieve this with a damped fixed-…
- arXiv cs.AIresearchHyperTaxa: Multiscale Genome Representations for Generalized Microbial Taxonomic Inference
arXiv:2610.25673v1 Announce Type: new Abstract: Reliable microbial taxonomic inference is challenged by the rapid growth of newly sequenced organisms and incomplete reference coverage. While alignment- and database-drive…
- arXiv cs.AIresearchLIRQuant: An Efficient Locally Isolated Rotation Quantization Method for Large Language Models
arXiv:2610.25675v1 Announce Type: new Abstract: Post-Training Quantization (PTQ) enables efficient deployment of Large Language Models (LLMs), yet activation outliers remain a key challenge for low-bit quantization. Exis…
- arXiv cs.AIresearchNeither Over- nor Under-Constrained: Evaluating LLM Personalization under a Constraint Qualification Audit
arXiv:2610.25676v1 Announce Type: new Abstract: Accurate evaluation of LLM personalization is hindered by current evaluation paradigms that over- or under-constrain the response space, yielding frequent false rejections…
- arXiv cs.AIresearchBitFlat: Error-Localized Bitstream Design for Robust Diffusion Image Compression
arXiv:2610.25682v1 Announce Type: new Abstract: Zero-shot diffusion image compression based on reverse channel coding achieves high perceptual quality at low bitrates without retraining, yet its bitstream remains vulnera…
- arXiv cs.AIresearchWhen Safety Fails Collectively: Robust Alignment Against Harmful Fine-Tuning via Joint Weight Perturbations
arXiv:2610.25683v1 Announce Type: new Abstract: Harmful fine-tuning can rapidly erode the safety alignment of large language models, creating substantial risks when aligned models are adapted to downstream tasks. Existin…
- arXiv cs.AIresearchCVaR-RMD: CVaR Selection Pressure for Robust Metaheuristic Discovery
arXiv:2610.25697v1 Announce Type: new Abstract: Large language models have recently been used to discover heuristics through iterative program search, offering a promising approach to automated algorithm design. However,…
- arXiv cs.AIresearchFrom Black Box to Controller: Inference-Time Steering of LLM via Feature Amplification
arXiv:2610.25700v1 Announce Type: new Abstract: Inference-time steering is crucial for deploying Large Language Models (LLMs) in real-world applications, where they must flexibly adjust behavior preferences. For instance…
- arXiv cs.AIresearchPFedQFT: Personalized Federated Quantum-Inspired Fine-Tuning for Medical Large Language Models
arXiv:2610.25712v1 Announce Type: new Abstract: Medical data remain fragmented across institutional silos because of privacy, governance, and ownership constraints. Federated learning enables medical institutions to adap…
- arXiv cs.AIresearchFine-Grained Units, Flat Decisions: Hierarchical Micro-Expert Pruning for Mixture-of-Experts Models
arXiv:2610.25717v1 Announce Type: new Abstract: Micro-expert pruning reduces the capacity of mixture-of-experts (MoE) models by removing coupled channels within routed experts. However, finer pruning granularity alone do…
- arXiv cs.AIresearchActTime: Action-Conditioned World Modeling for Time Series
arXiv:2610.25764v1 Announce Type: new Abstract: Time-series models increasingly condition on known future inputs. In controlled systems, some of these inputs are actions that can be changed, compared, and optimized befor…
- arXiv cs.AIresearchTRACE: Tracking Multimodal Clinical Trajectories for Inpatient Risk Prediction
arXiv:2610.25765v1 Announce Type: new Abstract: Accurate inpatient risk prediction requires modeling heterogeneous clinical evidence that accumulates irregularly over time. This challenge spans in-hospital mortality, pro…
- arXiv cs.AIresearchCan LLM User-Proxies Stand In for Humans in Reliance Measurement? Replicating Rel-A.I. with Simulated Users
arXiv:2610.25788v1 Announce Type: new Abstract: Can language-model user proxies reproduce human reliance on an AI answer across contexts? Published simulated-user studies do not establish validity for this partial-answer…
- arXiv cs.AIresearchSAR: Learning Global Item-Space Matching and Geometry-Aware Placement Ranking for Buffered Online 3D Bin Packing
arXiv:2610.25795v1 Announce Type: new Abstract: Online 3D bin packing with a bounded buffer requires irreversible item-selection and placement decisions under limited knowledge of future arrivals. Feasibility alone is in…
- arXiv cs.AIresearchWhen to Stop Asking: Factor Coverage Bias and LLM Determinability Judgments with Underspecified Information in Medicine
arXiv:2610.25819v1 Announce Type: new Abstract: Clinical decision-making often begins with underspecified information and requires not only gathering missing information but also knowing when to stop. This calls for judg…
- arXiv cs.AIresearchWhen Retrieval Partially Conflicts with Skills: Understanding Procedural Hallucination in Language Models
arXiv:2610.25827v1 Announce Type: new Abstract: Language models increasingly use skills that provide reusable instructions and procedures for task execution. After a relevant skill has been selected, the model may need t…
- arXiv cs.AIresearchNot Every Rollout Needs Distillation: Correctness-Filtered On-Policy Self-Distillation
arXiv:2610.25842v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) supervises student-generated trajectories with a reference-conditioned version of the same base model. We ask whether every trajectory su…
- arXiv cs.AIresearchA Cue's Hidden Price: Auditing Seller Influence on AI Shoppers
arXiv:2610.25843v1 Announce Type: new Abstract: Large language models are increasingly asked to shop on a user's behalf, which lets seller-controlled content such as badges, endorsements, and framing steer what they reco…
- arXiv cs.AIresearchMinimum-Cost Robust Evidence Certificates for Horn Reasoning under Correlated Source Failures
arXiv:2610.25865v1 Announce Type: new Abstract: A small justification for a rule-based conclusion may disappear when a source is withdrawn, and multiple derivations need not provide independent evidence. We study minimum…
- arXiv cs.AIresearchBenefit-Targeted Routing and Bounded Revision for LLM-Based Evidence Screening in Manufacturing Supplier Discovery
arXiv:2610.25867v1 Announce Type: new Abstract: An evidence-review system must allocate additional model judgments and control their use. We study these software decisions on 6,164 frozen manufacturing source-task pairs.…
- arXiv cs.AIresearchCODA: Cohort- and Drift-aware Foundation Model for Multimodal Clinical Reasoning
arXiv:2610.25871v1 Announce Type: new Abstract: Recent advances in multimodal foundation models have demonstrated strong performance across diverse clinical tasks. However, existing approaches predominantly operate at th…
- arXiv cs.AIresearchMotor compound fault diagnosis using an attention-based cross-talk architecture
arXiv:2610.25893v1 Announce Type: new Abstract: As physical AI increasingly relies on sensing to understand physical systems, reliable condition monitoring of industrial machinery has become more important. Motors are ub…
- arXiv cs.AIresearchGlobal probabilistic forecasting of air pollution
arXiv:2610.25941v1 Announce Type: new Abstract: Fine particulate matter (PM2.5) refers to particulate matter with an aerodynamic diameter of 2.5 micrometre or less. It comes from human activities, such as traffic and ind…
- arXiv cs.AIresearchOrderGPT: Generation Order as a Learnable Modeling Choice
arXiv:2610.25949v1 Announce Type: new Abstract: Generative modeling of discrete data has two popular paradigms: autoregressive models such as GPT, which factorize a sequence in a left-to-right (L2R) order, and masked dif…
- arXiv cs.AIresearchABACUS: Program-Guided Multistep Reasoning over Frozen Judgment Models
arXiv:2610.25963v1 Announce Type: new Abstract: Judgment models answer semantic questions by scoring a finite set of candidate outcomes. Can reusable programs turn these local decisions into powerful multistep reasoning?…
- arXiv cs.AIresearchGLOBE: Geometry-Led Online Behavioral Early Exit via Unit-Sphere Trajectories
arXiv:2610.25970v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning can improve problem solving, but models often keep generating after the correct answer is already reachable; the extra steps raise serving…
- arXiv cs.AIresearchOn the Benefits of Uncertainty Estimation for Species Classification: A Case Study on Optimizing Expert re-Annotation
arXiv:2610.25977v1 Announce Type: new Abstract: Automated identification of diatoms from microscopic images is a cornerstone of large-scale freshwater biomonitoring, yet the high diversity of species and strong class imb…
- arXiv cs.AIresearchGraphEvoResearcher: Graph-Aware Training-Free Self-Reflective Reasoning for Cost-Bounded Multi-Hop Question Answering
arXiv:2610.25978v1 Announce Type: new Abstract: Multi-hop question answering (QA) requires models to chain intermediate conclusions while remaining cost-aware. Existing training-free self-reflective protocols (generate -…
- arXiv cs.AIresearchTimesJev: Thinking Fast on Time Series with System 1 Sufficiency
arXiv:2610.26002v1 Announce Type: new Abstract: Large language models (LLMs) have recently shown strong potential for time series reasoning, while models that reason over time series are usually with long chains of thoug…