AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchEmbedding Geometry Shapes LLM Routing: Tracing Failure Modes and Improving Generalization
arXiv:2610.19198v1 Announce Type: new Abstract: With the growing number of LLMs available, selecting which one to use for a prompt has become a key question. A large part of the routing literature relies on semantically…
- arXiv cs.AIresearchPREFER: From Player Experience to Personalized Game Refinement
arXiv:2610.19213v1 Announce Type: new Abstract: Understanding how large language models (LLMs) evaluate gameplay experience is important for using their feedback to guide game design. However, overall ratings alone provi…
- arXiv cs.AIresearchREPORT-OPS: Hybrid Evidence-Guided Reporting with Human-Supervised Refinement for HPC
arXiv:2610.19222v1 Announce Type: new Abstract: Operational reporting for high-performance computing (HPC) facilities requires combining information from monitoring services, schedulers, incident and maintenance records,…
- arXiv cs.AIresearchDefinition Grounding in Theory-Level Auto-formalization
arXiv:2610.19223v1 Announce Type: new Abstract: LLM-based auto-formalization systems can now generate formally verified Lean code at unprecedented rates. However, a formalization advances mathematical research only when…
- arXiv cs.AIresearchWrite Once, Plan Many: Building Reusable Visual Planning Systems from One Scene
arXiv:2610.19234v1 Announce Type: new Abstract: Vision-language models (VLMs) can reason about visual planning tasks, but querying them for every instance incurs repeated inference cost and latency. Generating reusable p…
- arXiv cs.AIresearchBeyond ICD Codes: Semantic IDs for Generative Medical Coding
arXiv:2610.19301v1 Announce Type: new Abstract: Generative approaches to automatic International Classification of Diseases (ICD) coding use large language models (LLMs) to predict ICD codes from clinical documents. Howe…
- arXiv cs.AIresearchGeneralization Analysis of Biased Stochastic Gradient Methods for Minimax Problems
arXiv:2610.19322v1 Announce Type: new Abstract: Minimax optimization refers to a class of optimization problems with some variables to minimize and other variables to maximize. Biased stochastic gradient methods have sho…
- arXiv cs.AIresearchCan We Believe AI Peer Review?
arXiv:2610.19339v1 Announce Type: new Abstract: As large language models become widely used in scientific review and manuscript revision, AI scores are increasingly serving as feedback for assessing paper quality and gui…
- arXiv cs.AIresearchSteering Vectors for Competing Ethical Frameworks Are Interchangeable: A Competing-Target Test for Activation Steering
arXiv:2610.19359v1 Announce Type: new Abstract: Activation steering adds a vector to a model's hidden states to control a named behaviour. A gain on that behaviour, however, does not show that the vector controls it dire…
- arXiv cs.AIresearchDANGERLENS: Probing Dangerous Representations Hidden by Safety Alignment
arXiv:2610.19365v1 Announce Type: new Abstract: Safety alignment teaches language models to refuse harmful requests, yet the internal accessibility of hazardous concepts can retain traces of earlier training. Can checkpo…
- arXiv cs.AIresearchBeyond the Trial-and-Error Loop: Hybrid Projection and Automated Tuning for Distributed Training
arXiv:2610.19423v1 Announce Type: new Abstract: Configuring distributed LLM training across hundreds of GPUs requires jointly optimizing tensor, pipeline, expert, context, and data parallelism alongside micro-batch size,…
- arXiv cs.AIresearchRecognition Is Not Grounding: Benchmarking Multilingual Financial Evidence Extraction
arXiv:2610.19432v2 Announce Type: replace Abstract: Comparing financial disclosures across languages requires identifying shared accounting concepts and recovering all numerical facts that express them. Existing benchmar…
- arXiv cs.AIresearchIntegrating Simulation and Chain-of-thought Reasoning in Multimodal-Language Models For Physical Reasoning
arXiv:2610.19441v1 Announce Type: new Abstract: Multimodal language models often need to reason about the physical dynamics of the world in order to answer questions appropriately. Various methods have emerged for solvin…
- arXiv cs.AIresearchxLightGCN: Explainable Graph Neural Network Based Recommendation Systems with Most Influential Peers
arXiv:2610.19446v1 Announce Type: new Abstract: Recommendation systems, systems that predict a user's preference for items in order to curate that user's feed, are a ubiquitous and central feature in online social networ…
- arXiv cs.AIresearchAre You Reading the Right Thing? Rethinking Internal Signals for Prompt-Injection Detection
arXiv:2610.19452v1 Announce Type: new Abstract: Several recent studies propose detecting prompt injections from a language model's internal state. However, these methods do not measure a single security property. Residua…
- arXiv cs.AIresearchVLRLA: Deployment-Time Adaptation of Vision-Language-Action Models through In-Context Reinforcement Learning
arXiv:2610.19517v1 Announce Type: new Abstract: Vision-language-action (VLA) models connect visual perception and language understanding with robot control, providing a foundation for instruction-conditioned manipulation…
- arXiv cs.AIresearchRevisiting Triplet Contrastive Learning for 3D Shape Understanding
arXiv:2610.19519v1 Announce Type: new Abstract: In multi-modal 3D pre-training, aligning 3D shapes, their 2D image counterparts, and language descriptions has made tremendous progress. However, we find that supervision f…
- arXiv cs.AIresearchDivergence-Routed On-Policy Distillation for Language Models
arXiv:2610.19572v1 Announce Type: new Abstract: On-policy distillation (OPD) transfers knowledge from a teacher model to a student along student-generated trajectories, typically by minimizing the reverse Kullback-Leible…
- arXiv cs.AIresearchWho's Behind the Text? Robust Model Authorship Attribution
arXiv:2610.19583v1 Announce Type: new Abstract: As large language models (LLMs) are increasingly used to generate text, their misuse has become a growing societal problem, with implications for academic misconduct, model…
- arXiv cs.AIresearchAre AI Decision Models Biased? A Case Study of Jev
arXiv:2610.19619v1 Announce Type: new Abstract: Do AI models designed for fast, structured decisions exhibit the cognitive and social biases observed in human judgment? We evaluate Jev, a model described as a System One…
- arXiv cs.AIresearchLearning When to Trust Medical Knowledge: Regression-Safe Routing for Clinical Reasoning
arXiv:2610.19663v1 Announce Type: new Abstract: Clinical reasoning LLMs need external medical knowledge, but retrieved evidence and reasoning chains can be outdated, poisoned, or hallucinated. Harmful knowledge can cause…
- arXiv cs.AIresearchReFormBench: Benchmarking the Stability of Response-Conditioned Format Correction in Large Language Models
arXiv:2610.19693v1 Announce Type: new Abstract: Reliable use of large language models requires responses to satisfy formatting requirements across changing instructions. Standard compliance scores do not distinguish pres…
- arXiv cs.AIresearchRuntime Monitoring of Supermartingale Certificates
arXiv:2610.19701v1 Announce Type: new Abstract: Supermartingale certificates formally guarantee temporal properties of discrete and continuous-time stochastic dynamical systems. However, these guarantees rely on a mathem…
- arXiv cs.AIresearchLearning Individual Scientific Taste with Privileged Hindsight Distillation
arXiv:2610.19719v1 Announce Type: new Abstract: Scientific taste shapes which research directions researchers pursue, and it varies across individuals. Existing work evaluates research directions along multiple evaluatio…
- arXiv cs.AIresearchArticulation Upper Bounds for Language Models
arXiv:2610.19721v1 Announce Type: new Abstract: Language captures only a subset of the perceptions, intuitions and intents that form meaning between speakers. What, then, are the limits of language? We take steps towards…
- arXiv cs.AIresearchLabOPBench: A Multimodal Benchmark for Evidence-Grounded Laboratory Decision Making
arXiv:2610.19780v1 Announce Type: new Abstract: Laboratory experiments involve more than following predefined protocols; they require adaptive decision-making as the experimental state changes. Existing scientific benchm…
- arXiv cs.AIresearchSparse Transformer: a Controlled Testbed for Mechanistic Interpretability
arXiv:2610.19907v1 Announce Type: new Abstract: Mechanistic interpretability aims to decompose neural network computations into human-understandable features, yet progress is limited by the absence of ground-truth units…
- arXiv cs.AIresearchWhen Protection Overrides Consequences: Understanding Protection Mode in Large Language Model Decision-Making
arXiv:2610.19948v1 Announce Type: new Abstract: Frontier language models excel at complex tasks yet may prioritize protection over larger physical consequences in simple, urgent decisions. GPT-5.6 Sol, for example, moves…
- arXiv cs.AIresearchBeyond Correctness: Compute-Responsive Data Selection for Reasoning Model Post-Training
arXiv:2610.20057v1 Announce Type: new Abstract: Reasoning language models increasingly rely on test-time compute scaling, yet most existing work studies how to allocate inference computation while treating the model itse…
- arXiv cs.AIresearchDesigning Textual Interfaces To Enhance Llm Reasoning On Audio-Visual Question Answering
arXiv:2610.20085v1 Announce Type: new Abstract: Automatic prompt optimization (APO) has shown strong effectiveness for text-based LLMs and LLM systems, but its application to audio-visual understanding systems requires s…
- arXiv cs.AIresearchLLM-Based Policy Compliance Verification of Loan Applications
arXiv:2610.20155v1 Announce Type: new Abstract: Banks rely on extensive internal policy frameworks to ensure that corporate lending decisions comply with regulatory requirements and internal risk standards. Reviewing loa…
- arXiv cs.AIresearchValue-Guided Repulsion for Offline Reinforcement Learning
arXiv:2610.20190v1 Announce Type: new Abstract: Drifting models have emerged as a highly efficient and stable paradigm for policy design in offline reinforcement learning (RL). However, standard drift policies determine…
- arXiv cs.AIresearchRobust Counterfactuals for Markov Decision Processes under Statistical and Causal Uncertainties
arXiv:2610.20210v1 Announce Type: new Abstract: Counterfactual reasoning (asking what would have happened, had a decision been different) is a powerful tool for identifying causes of failure and explaining the behaviour…
- arXiv cs.AIresearchDesign Space Exploration of Reinforcement Learning Configurations for Post-Legalization Placement
arXiv:2610.20224v1 Announce Type: new Abstract: Graph Neural Networks (GNNs) and Reinforcement Learning (RL) have recently been explored for VLSI placement optimization, but their effectiveness for post-legalization plac…
- arXiv cs.AIresearchArgument Graph Uncertainty: Quantifying Uncertainty from the Logical Structure of Reasoning Chains
arXiv:2610.20328v1 Announce Type: new Abstract: Reliable uncertainty estimation is essential for deploying large language models in high-stakes reasoning tasks, yet most existing approaches rely on expensive sampling str…
- arXiv cs.AIresearchLifted State Policy Optimization: Internalizing Self-Correction through Transition Geometry
arXiv:2610.20350v1 Announce Type: new Abstract: Preference optimization ranks final answers and trajectory RL maximizes aggregate returns, yet neither credits the local change that turns a flawed reasoning state into a b…
- arXiv cs.AIresearchSafeAdapt-Poker: Counterfactual Risk Calibration for Budgeted Adaptation under Opponent Shift
arXiv:2610.20473v2 Announce Type: replace Abstract: Opponent histories can mislead after behavioral shifts. SafeAdapt-Poker formulates budgeted adaptation under opponent shift as a dual-counterfactual sequential-control…
- arXiv cs.AIresearchLotus: Learning to Understand Time-series by Learning to Forecast
arXiv:2610.18877v1 Announce Type: cross Abstract: Time-series language models typically align numerical signals with language using labels, captions, or question-answer pairs. Although such supervision encourages linguis…
- arXiv cs.AIresearchInverse Reasoning Models for Efficient Post-Training and Verification from Answer-Only Data
arXiv:2610.18878v1 Announce Type: cross Abstract: Post-training reasoning LLMs often requires making a tradeoff between costly supervision and compute inefficiency: supervised finetuning typically requires reasoning trac…
- arXiv cs.AIresearchDo Vision CIL Methods Transfer to Protein? A Diagnostic Benchmark of Frozen Protein Encoders
arXiv:2610.18879v2 Announce Type: replace-cross Abstract: Protein function annotation operates over an expanding label space, making class-incremental learning (CIL) a natural framework for incorporating newly discovered…
- arXiv cs.AIresearchToken Utility for On-Policy Distillation via Local Trust-Region Optimization
arXiv:2610.18883v1 Announce Type: cross Abstract: Token-selection methods for on-policy distillation (OPD) prioritize student-visited positions using diverse signals, including uncertainty, teacher-student disagreement,…
- arXiv cs.AIresearchPersonalising Deformable Image Registration with Clinical Context and Deformation History for Adaptive Radiotherapy
arXiv:2610.18926v1 Announce Type: cross Abstract: Adaptive radiotherapy must account for anatomical changes during treatment, but doing so depends on reliably mapping targets, organs at risk and dose across serial images…
- arXiv cs.AIresearchNon-Abelian quantum Hall states at strong Landau level mixing: finding anyons with first-principles AI
arXiv:2610.18967v1 Announce Type: cross Abstract: Non-Abelian anyons, strange particles forming the basis of fault-tolerant topological quantum computing, are predicted to emerge in fractional quantum Hall states at even…
- arXiv cs.AIresearchEvaluating Large Language Models for Multilingual Abstractive Summarization: A Comprehensive Study of English and Bengali
arXiv:2610.19003v1 Announce Type: cross Abstract: Abstractive summarization remains challenging across languages, where linguistic quality and factual consistency may diverge, particularly in low-resource settings. We pr…
- arXiv cs.AIresearchScaling Cheap Synthetic Pretraining Improves Indoor Radio-Map Prediction
arXiv:2610.19015v1 Announce Type: cross Abstract: Real measurements are too limited to train an accurate indoor pathloss predictor that works across a wide range of buildings. Large-scale synthetic pretraining offers a w…
- arXiv cs.AIresearchWho Should We Listen to More? Welfare-Aware Preference Acquisition for Pluralistic Alignment
arXiv:2610.19035v1 Announce Type: cross Abstract: Pluralistic alignment needs to accommodate diverse preferences under limited human feedback. Using AI simulators to label preference comparisons reduces annotation costs,…
- arXiv cs.AIresearchLoss Curvature as a Sampling-Free Proxy for Memorization in Diffusion Models
arXiv:2610.19042v1 Announce Type: cross Abstract: Understanding the mechanism behind why diffusion models memorize their training data remains an open question in generative modeling. In this paper, we examine this quest…
- arXiv cs.AIresearchGraph Coarsening Enables Escaping Local Optima in Compute-Constrained Nonconvex MAX-CUT
arXiv:2610.19043v1 Announce Type: cross Abstract: Because its binary optimization landscape contains exponentially many local optima, MAX-CUT is highly sensitive to initialization. We study Particle Polynomial Relaxation…
- arXiv cs.AIresearchHow Inner Optimizers Shape Replica Scaling
arXiv:2610.19064v1 Announce Type: cross Abstract: Training large language models (LLMs) across geographically distributed and poorly interconnected compute pools or data centers has become a topic of significant interest…
- arXiv cs.AIresearchRevisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap
arXiv:2610.19065v1 Announce Type: cross Abstract: Generative retrieval (GR) has emerged as a strong candidate generator for sequential recommendation as language models and semantic-item identifiers have improved. Howeve…