AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchWhat Pairwise Evaluation Misses: Reward Scores across Prompts
arXiv:2610.31090v1 Announce Type: cross Abstract: Reward models score responses for preference optimization, filtering and ranking, and pooling scores across prompts needs them to be comparable. Yet pairwise evaluation c…
- arXiv cs.AIresearchA Grammar of French Civil Cassation: A Ground-Level Dataset and Tool
arXiv:2610.31116v1 Announce Type: cross Abstract: This paper presents a dataset of 121,536 ground-reasons pairs across 86,464 rulings of the five civil chambers of the French Court of Cassation (2016-2025). An appeal on…
- arXiv cs.AIresearchMixing Matters: Evidence Position Bias Across Sequence Mixers in Long-Context Question Answering
arXiv:2610.31135v1 Announce Type: cross Abstract: Long-context models often use answer-bearing evidence less effectively when it appears away from a prompt boundary. We compare this position bias across Transformer, stat…
- arXiv cs.AIresearchPriMa-Causa: Causal Prescriptive Maintenance
arXiv:2610.31147v1 Announce Type: cross Abstract: Automated manufacturing requires data-driven decisions under uncertainty and sparse observational records, and unplanned downtime is a dominant cost of getting them wrong…
- arXiv cs.AIresearchPrompt Sensitivity in Tool-Augmented Model Comparisons
arXiv:2610.31148v1 Announce Type: cross Abstract: Language models are compared on question-answering benchmarks that pose each question in one wording, and a reworded question can change which of two models scores higher…
- arXiv cs.AIresearchNetwork topology shapes consensus and vulnerability in LLM populations
arXiv:2610.31165v1 Announce Type: cross Abstract: As large language models (LLMs) are deployed in interacting populations, their collective behaviour depends not only on the models but also on who interacts with whom. St…
- arXiv cs.AIresearchPrompting Strategy Effects Depend on the Judge: A Multilingual Study of LLM-as-a-Judge Robustness
arXiv:2610.31172v1 Announce Type: cross Abstract: LLM-as-a-Judge evaluation is increasingly applied to multilingual tasks, yet prompting strategies such as Chain-of-Thought and Role Conditioning are recommended based sol…
- arXiv cs.AIresearchWhen More Metadata Misleads: Probing Field-Level Grounding in LLM-Generated Museum Descriptions
arXiv:2610.31193v1 Announce Type: cross Abstract: Metadata-conditioned generation is often framed as a data-to-text task in which additional input fields should improve grounding. Museum description generation challenges…
- arXiv cs.AIresearchPredicting Music Valence and Arousal via Textual and Multimodal In-Context Learning in Large Language Models
arXiv:2610.31202v1 Announce Type: cross Abstract: While arousal can be reliably predicted from musical audio, emotional valence remains difficult to model due to its subjective and context-dependent nature. In this work,…
- arXiv cs.AIresearchQICOM: Open-Source Software for Reproducible Image Quality Evaluation, Blinded Assessment, and Radiomic Comparison
arXiv:2610.31218v1 Announce Type: cross Abstract: QICOM (Quality Image COMparator) is open-source software for quantitative image comparison, paired radiomics, and blinded assessment. It saves processing settings, image…
- arXiv cs.AIresearchCan Entanglement-Induced Feature-Map Designs Improve Projected Quantum Kernels?
arXiv:2610.31260v1 Announce Type: cross Abstract: Projected quantum kernels (PQKs) have emerged as a notable approach to circumvent the hardness of machine learning tasks arising from the exponential dimensionality of th…
- arXiv cs.AIresearchNotation3: Graph Terms for RDF
arXiv:2610.31266v1 Announce Type: cross Abstract: Notation3 Logic was introduced as an extension of RDF allowing the user to express rules on top of triples and to make statements about graphs. The original proposal of t…
- arXiv cs.AIresearchCoWriter: Guiding Methodological and Synergetic Human-LLM Co-Authoring
arXiv:2610.31277v1 Announce Type: cross Abstract: While LLMs have transformed text generation, standard chat-and-canvas interfaces often reduce human authoring to quite passive supervision, causing review fatigue, over-r…
- arXiv cs.AIresearchRecallGS: Geometry-Grounded Recall for Online Semantic Gaussian Splatting
arXiv:2610.31293v1 Announce Type: cross Abstract: Online semantic 3D Gaussian Splatting (3DGS) enables open-vocabulary scene understanding from streaming observations. However, existing methods recover geometric context…
- arXiv cs.AIresearchWhen LLMs Forget What They Believe: Characterizing State Integrity Collapse in Long-Horizon Dialogues
arXiv:2610.31314v1 Announce Type: cross Abstract: Large language models can repeat an instruction while failing to keep obeying it. We study this recall-compliance dissociation in long-horizon dialogue and define State I…
- arXiv cs.AIresearchSurveyGraph: Generating Academic Surveys via Methods-Entity Knowledge Graphs
arXiv:2610.31317v1 Announce Type: cross Abstract: Most automated survey generation systems retrieve papers as flat documents, limiting their ability to capture research structure. We present SurveyGraph, a framework buil…
- arXiv cs.AIresearchHyperreduction accelerates finite element model updating for the discovery of neural materials
arXiv:2610.31322v1 Announce Type: cross Abstract: Finite element model updating is among the most general approaches to discover constitutive models from structural experiments: since the displacement field is computed r…
- arXiv cs.AIresearchBridging the Data Gap: Source-Informed Task-Space Transport for Few-Shot Diffusion
arXiv:2610.31323v1 Announce Type: cross Abstract: Many deployments require a model to move rapidly from a data-rich source to a related target observed only sparsely. This is especially consequential where target data ar…
- arXiv cs.AIresearchAI Risk Ontologies for the EU AI Act Compliance: AIRO, VAIR, and their Integration with DPV
arXiv:2610.31380v1 Announce Type: cross Abstract: As the use of AI, and consequently its harms, becomes more prevalent, the global discourse around AI has become increasingly risk-aware. With new risks emerging more freq…
- arXiv cs.AIresearchAuditing the Hand-Written Inputs of Learned Access Control
arXiv:2610.31388v1 Announce Type: cross Abstract: Learned access control is typically deployed behind two hand-built artefacts: a hard-deny set enforced by a small trusted shield, and a region map that decides which requ…
- arXiv cs.AIresearchHERD: Harness-Evolving Recursive Distillation for Continual Learning
arXiv:2610.31401v1 Announce Type: cross Abstract: Continual learning requires a model to keep acquiring new skills and knowledge without compromising its existing capabilities. Parametric methods suffer from catastrophic…
- arXiv cs.AIresearchSame Entity, Different Script: Diagnosing Cross-Script Entity Matching Failures in Language Models
arXiv:2610.31407v1 Announce Type: cross Abstract: Multilingual language models can give different identity judgments when the same people's names are written in different scripts. We study this effect using 601 audited i…
- arXiv cs.AIresearchMajor Revision, Minor Accountability: LLMs as Reviewers of Published Articles in Top-Tier Journals
arXiv:2610.31414v1 Announce Type: cross Abstract: With the growth of scientific output, the peer review system is approaching a breaking point: reviewers are increasingly hard to secure, and turnaround times lengthen acc…
- arXiv cs.AIresearchPost-Hoc Feature Selection Layer for Neural Networks Interpretability
arXiv:2610.31420v1 Announce Type: cross Abstract: The interpretability of complex neural networks remains a critical challenge, especially for models already deployed in high-stakes domains. To address this, we adapted t…
- arXiv cs.AIresearchHow Depth Growth and Looping Shape In-Context Learning in Language Models
arXiv:2610.31421v1 Announce Type: cross Abstract: Looped and depth-grown language models reuse computations across depth and have attracted interest for their reasoning capabilities. Building on prior suggestive evidence…
- arXiv cs.AIresearchFocusVLA: Future-Intent Guided Dynamic View Routing for Vision-Language-Action Models
arXiv:2610.31423v1 Announce Type: cross Abstract: Vision-language-action (VLA) models have become an important paradigm for robotic manipulation by unifying visual perception, language understanding, and action generatio…
- arXiv cs.AIresearchPredictable performance is suboptimal, and what it means for model routing
arXiv:2610.31428v1 Announce Type: cross Abstract: Large language models (LLMs) differ in capabilities and costs, motivating routing methods that adaptively choose a model for each query. These methods rely on performance…
- arXiv cs.AIresearchMoG-VLM: Efficient Video Anomaly Detection through Motion-Guided Vision-Language Models
arXiv:2610.31478v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have shown strong potential for video anomaly detection, but dense frame-level inference makes them impractical for real-time deployment. Ex…
- arXiv cs.AIresearchSABLE: Scalable Affine Bayesian Latent-space Exploration
arXiv:2610.31495v2 Announce Type: replace-cross Abstract: Complex objects such as engineering designs and proteins are often difficult to optimise because their design spaces exhibit high-dimensional structure and object…
- arXiv cs.AIresearchThe Geometry of Metaphor: Probing and Steering Metaphor Representations in Large Language Models
arXiv:2610.31520v1 Announce Type: cross Abstract: Metaphor is a crucial aspect of human cognition. As such, we would like to understand how it is represented in large language models (LLMs) and how it might be controlled…
- arXiv cs.AIresearchWhen to Follow the Teacher: Advantage-Based On-Policy Distillation
arXiv:2610.31563v1 Announce Type: cross Abstract: Online Policy Distillation (OPD) often outperforms offline methods such as Supervised Fine-Tuning (SFT) by training on trajectories sampled directly from the student. How…
- arXiv cs.AIresearchSocial-HUMINT: An LLM-Augmented Unified Framework for Measuring Social Media Exposure
arXiv:2610.31573v1 Announce Type: cross Abstract: In the post-API landscape, social-media profiling has become inexpensive and automated: Large Language Models (LLMs) can extract structured information from unstructured…
- arXiv cs.AIresearchWould an LLM Match into a Residency? Uncertainty, Reasoning and Image Use of Open Language Models on the Spanish MIR Examination
arXiv:2610.31586v1 Announce Type: cross Abstract: Accuracy on medical examinations is common evidence offered for the clinical use of language models, yet a health system that intends to run an open model on its own hard…
- arXiv cs.AIresearchVERA: A Responsible AI Dashboard for Assessing LLM Deployments in Light of the EU AI Act
arXiv:2610.31588v1 Announce Type: cross Abstract: Institutions that deploy large language models in regulated settings often need technical evidence of responsible AI practices before approving a deployment. COMPL-AI pro…
- arXiv cs.AIresearchHow Ground Truth Shapes Retrieval Scores: Evidence-Aware Evaluation of a Retrieval-Augmented Generation System for Breast Lesions
arXiv:2610.31606v2 Announce Type: replace-cross Abstract: Purpose: To describe an evidence-aware retrieval evaluation protocol for a clinical retrieval-augmented generation (RAG) system over breast lesion management lite…
- arXiv cs.AIresearchCausal Chain of Thought: Mitigating Attention Leakage via Context Isolation
arXiv:2610.31639v1 Announce Type: cross Abstract: Chain-of-thought (CoT) prompting elicits step-by-step rationales that appear to explain model predictions, but there is no guarantee these steps causally drive the final…
- arXiv cs.AIresearchLow-Bit N-Gram Embeddings for Language Models
arXiv:2610.31646v1 Announce Type: cross Abstract: Many recent open-weight language models use n-gram embedding tables to improve performance. These tables store learned representations of short token sequences, such as p…
- arXiv cs.AIresearchCost-Aware Abstention for LLM-Based IT Ticket Classification
arXiv:2610.31657v1 Announce Type: cross Abstract: Wrong automatic routing of an IT support ticket can cost more than human triage. We evaluate cost-aware abstention as an offline, cost-conditional readiness assessment: s…
- arXiv cs.AIresearchAnchoring Distributed Agency: Institutional Metaphors for Designing Large-Scale AI and AI Techniques for Running Institutions
arXiv:1803.03407v3 Announce Type: replace Abstract: Artificial Intelligence (AI) started out with the ambition to reproduce the human mind, but as soon as the sheer scale of that ambition became apparent, it retreated in…
- arXiv cs.AIresearchRisk-Aware Alignment Must Incorporate Reward Uncertainty
arXiv:2410.23726v2 Announce Type: replace Abstract: We consider the problem of policy optimization with noisy rewards, where the policy is constrained to be in the neighborhood of a reference policy.We study the risk of…
- arXiv cs.AIresearchGroup Contrastive Policy Optimization: Learning Beyond the Self-Exploration Limit with Golden Answers
arXiv:2510.07790v2 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a powerful paradigm for improving the reasoning abilities of large language models, yet group-based met…
- arXiv cs.AIresearchHow to Teach Large Multimodal Models New Skills
arXiv:2510.08564v3 Announce Type: replace Abstract: How can we teach large multimodal models (LMMs) new skills without erasing prior abilities? We study sequential fine-tuning on five target skills while monitoring gener…
- arXiv cs.AIresearchMENTOR: A Metacognition-Driven Self-Evolution Framework for Uncovering and Mitigating Implicit Domain Risks in LLMs
arXiv:2511.07107v5 Announce Type: replace Abstract: Ensuring the safety of Large Language Models (LLMs) is critical for real-world deployment. However, current safety measures often fail to address implicit, domain-speci…
- arXiv cs.AIresearchGFlowPO: Generative Flow Network as a Language Model Prompt Optimizer
arXiv:2602.03358v2 Announce Type: replace Abstract: Finding effective prompts for language models (LMs) is critical yet notoriously difficult: the prompt space is combinatorially large, rewards are sparse due to expensiv…
- arXiv cs.AIresearchSleepMaMi: A Universal Sleep Foundation Model for Integrating Macro- and Micro-structures
arXiv:2602.07628v3 Announce Type: replace Abstract: While the shift toward unified foundation models has revolutionized many deep learning domains, sleep medicine remains largely restricted to task-specific models that f…
- arXiv cs.AIresearchSci-CoE: Co-evolving Scientific Reasoning LLMs via Geometric Consensus with Sparse Supervision
arXiv:2602.12164v2 Announce Type: replace Abstract: Large language models (LLMs) have demonstrated exceptional reasoning capabilities, and co-evolving paradigms have shown promising results in domains with verifiable rew…
- arXiv cs.AIresearchProjecting Latent RL Actions: Towards Generalizable and Scalable Graph Combinatorial Optimization
arXiv:2605.19721v2 Announce Type: replace Abstract: Graph combinatorial optimization (GCO) has attracted growing interest, as many NP-hard problems naturally admit graph formulations, yet their combinatorial explosion re…
- arXiv cs.AIresearchXetrieval: Explaining Dense Retrieval through Feature Elicitation
arXiv:2605.29507v2 Announce Type: replace Abstract: Dense retrieval enables effective semantic matching, yet the semantic factors underlying its relevance scores remain opaque. Sparse autoencoders (SAEs) support feature-…
- arXiv cs.AIresearchDGA$_2$D: Directed Graph-Guided Automated Algorithm Design with Large Language Models
arXiv:2608.00700v2 Announce Type: replace Abstract: The rapid development of Large Language Models (LLMs) has opened new avenues for Automated Heuristic Design (AHD) for solving NP-hard combinatorial optimization problem…
- arXiv cs.AIresearchWhen Self-Consistency Backfires: Majority Vote Hurts the Majority of Hard Science Problems for Small LLMs
arXiv:2608.11403v3 Announce Type: replace Abstract: Majority-vote self-consistency (SC) lowers per-problem accuracy on most GPQA Diamond problems for small instruction-tuned models: on 56.6% of problems for Qwen2.5-7B an…