AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchGeoReader: Learning Latent Geometry for Spatial Reasoning in Vision-Language Models
arXiv:2610.18694v1 Announce Type: cross Abstract: Vision-language models (VLMs) struggle with metric and multi-view relationships that remain implicit in RGB observations. Geometry foundation models recover this structur…
- arXiv cs.AIresearchUniW(A)M: A Unified Mixture-of-Experts Framework for World and World-Action Models
arXiv:2610.18697v2 Announce Type: replace-cross Abstract: World models (WMs) and world-action models (WAMs) focus on future world prediction and executable action generation, respectively. Although these models learn sha…
- arXiv cs.AIresearchSteertuning Policies from a Few Human Videos
arXiv:2610.18711v1 Announce Type: cross Abstract: Current recipes for robot learning require massive amounts of robot demonstrations to obtain robust and adaptive policies. In this work, we introduce Steertuning, a syste…
- arXiv cs.AIresearchA Rate Transfer Method: Convergence Rates in Stochastic Gradient Descent via Stochastic Calculus
arXiv:2610.18739v1 Announce Type: cross Abstract: We develop a rate-transfer method that converts finite-time Lyapunov estimates for an associated stochastic differential equation (SDE) into finite-time convergence bound…
- arXiv cs.AIresearchSolBehave: Evaluating the Behavioural Correctness of LLM-Generated Smart Contracts
arXiv:2610.18745v1 Announce Type: cross Abstract: Large language models increasingly generate Solidity smart contracts that, once deployed, can never be patched, making a single latent fault permanent and exploitable. Ye…
- arXiv cs.AIresearchPreciseCoder: Faithful and Generalizable Debugging Beyond Unit-Test Rewards
arXiv:2610.18755v1 Announce Type: cross Abstract: Reinforcement learning (RL) from unit tests is the common recipe for post-training code models, but a unit test only asks whether the final program passes. We analyze wha…
- arXiv cs.AIresearchWhere Do LLM Ensemble Gains Come From? Decomposing Candidate Value, Diversity, and Aggregation
arXiv:2610.18786v2 Announce Type: replace-cross Abstract: LLM ensemble gains are often measured against a single baseline, but the value attributed to additional candidates can change with the comparison used. We study a…
- arXiv cs.AIresearchBytes to Threats: Byte-Sequence Transformer For Malware Detection
arXiv:2610.18787v1 Announce Type: cross Abstract: Existing malware detection approaches heavily rely on handcrafted features derived from static or dynamic analysis of the executable files. These methods require speciali…
- arXiv cs.AIresearchBack-Translation Still Matters: Leveraging Monolingual Data for Low-Resource Machine Translation with LLMs
arXiv:2610.18818v3 Announce Type: replace-cross Abstract: Pre-trained Large Language Model (LLMs) are effective on many tasks in zero-shot and few-shot settings. For machine translation of low-resource languages, the obv…
- arXiv cs.AIresearchFast Rates for Offline Q-Learning with KL Regularization
arXiv:2610.18839v1 Announce Type: cross Abstract: This paper investigates offline Q-learning for finite-horizon Markov decision processes with Kullback-Leibler (KL) regularization. Recent studies have shown that KL regul…
- arXiv cs.AIresearchLIBERO-RIG: A Hardware-Setup Robustness Benchmark for Vision-Language-Action Models
arXiv:2610.18843v1 Announce Type: cross Abstract: Robot policies are expected to generalize across deployment environments, including variations in the robot itself: cameras are bumped and remounted, calibrations drift,…
- arXiv cs.AIresearchTwo Clocks: Why Larger Rollout Batches Learn Slower but Fail on Schedule in Reinforcement Learning with Verifiable Rewards
arXiv:2610.18853v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) relies on costly model-generated responses, motivating larger rollout batches and square-root learning-rate scaling.…
- arXiv cs.AIresearchDoubtLess: Training-Free Contrastive Decoding for Efficient Reasoning
arXiv:2610.18862v1 Announce Type: cross Abstract: Reasoning models achieve strong performance through long chains of thought (CoT), but often spend more tokens than needed to solve a problem. Existing ways to shorten the…
- arXiv cs.AIresearchWhat Matters in Quality-Aware VLA Pretraining: A Benchmark and Empirical Study
arXiv:2610.18869v1 Announce Type: cross Abstract: How should demonstration quality be used during vision-language-action (VLA) pretraining to improve performance after target-task fine-tuning? We compare weighting, filte…
- arXiv cs.AIresearchIdentification of Entailment and Contradiction Relations for Natural Language Argumentation: A Neurosymbolic Approach
arXiv:2405.01259v2 Announce Type: replace Abstract: Argumentation is an important cognitive activity that involves the scrutiny of arguments and counterarguments. When given an argument, it is desirable that its premises…
- arXiv cs.AIresearchOvercoming a Learning Bottleneck in Impartial Games: A Circuit Complexity Approach
arXiv:2411.06403v4 Announce Type: replace Abstract: Why can neural networks struggle to learn optimal play in a game with a simple, efficient solution algorithm? Motivated by empirical difficulties with parity in imparti…
- arXiv cs.AIresearchRobustness and Cybersecurity in the EU Artificial Intelligence Act
arXiv:2502.16184v4 Announce Type: replace Abstract: The EU Artificial Intelligence Act (AIA) establishes different legal principles for different types of AI systems. While prior work has sought to clarify some of these…
- arXiv cs.AIresearchPhysics: Benchmarking Foundation Models on University-Level Physics Problem Solving
arXiv:2503.21821v2 Announce Type: replace Abstract: We introduce Physics, a benchmark for evaluating foundation models on advanced physics problem solving. It contains 1034 annotated, open-ended problems from PhD-qualify…
- arXiv cs.AIresearchLLM-BabyBench: Can Language Models Plan in Worlds They Can Simulate?
arXiv:2505.12135v3 Announce Type: replace Abstract: Interactive benchmarks for LLMs usually report a single metric, even though succeeding in them requires models to show several different skills. While those benchmarks…
- arXiv cs.AIresearchDecomposability-Guided Cooperative Coevolution for Large-Scale Itinerary Planning
arXiv:2506.06121v3 Announce Type: replace Abstract: Large-scale itinerary planning is a variant of the traveling salesman problem, aiming to determine an optimal path that maximizes the collected point-of-interest (POI)…
- arXiv cs.AIresearchCite Pretrain: Retrieval-Free Knowledge Attribution for Large Language Models
arXiv:2506.17585v4 Announce Type: replace Abstract: Large language models can answer questions correctly and still hallucinate their citations. Most citation systems address this by retrieving documents at inference time…
- arXiv cs.AIresearchToward a Universal Characterization of Sudoku Puzzle Difficulty
arXiv:2507.21137v2 Announce Type: replace Abstract: In this paper we try to answer the question "What constitutes Sudoku difficulty rating across different Sudoku websites?" Using two distinct methods that can both solve…
- arXiv cs.AIresearchOPTIC-ER: A Reinforcement Learning Framework for Real-Time Emergency Response and Equitable Resource Allocation in Underserved African Communities
arXiv:2508.12943v3 Announce Type: replace Abstract: Public service systems in many African regions suffer from delayed emergency response and spatial inequity, causing avoidable suffering. This paper introduces OPTIC-ER,…
- arXiv cs.AIresearchLIFT: Likelihood-Based Intervention Faithfulness Training
arXiv:2509.13334v2 Announce Type: replace Abstract: Chain-of-thought (CoT) reasoning improves LLM accuracy on complex tasks, but CoTs are often unfaithful; that is, the model does not rely on the stated reasoning steps t…
- arXiv cs.AIresearchCompositional Machine Design as Program Synthesis with LLMs
arXiv:2510.14980v4 Announce Type: replace Abstract: Large language models (LLMs) have shown strong abilities in writing and revising programs, yet many program-synthesis benchmarks still evaluate programs in symbolic or…
- arXiv cs.AIresearche1: Learning Adaptive Control of Reasoning Effort
arXiv:2510.27042v3 Announce Type: replace Abstract: Increasing the thinking budget of AI models can significantly improve accuracy, but not all questions warrant the same amount of reasoning. Users may prefer to allocate…
- arXiv cs.AIresearchAlignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg Equilibria
arXiv:2511.17937v2 Announce Type: replace Abstract: Alignment faking is a form of strategic deception in AI in which models selectively comply with training objectives when they infer that they are in training, while pre…
- arXiv cs.AIresearchAlignMerge - Alignment-Preserving Large Language Model Merging via Fisher-Guided Geometric Constraints
arXiv:2512.16245v2 Announce Type: replace Abstract: Merging large language models (LLMs) is a practical way to compose capabilities from multiple fine-tuned checkpoints without retraining. Yet standard schemes (linear we…
- arXiv cs.AIresearchROI-Reasoning: Rational Optimization for Inference via Pre-Computation Meta-Cognition
arXiv:2601.03822v2 Announce Type: replace Abstract: Large language models (LLMs) can achieve strong reasoning performance with sufficient computation, but they do not inherently know how much computation a task requires.…
- arXiv cs.AIresearchGDEPO: Group Dual-dynamic and Equal-right Advantage Policy Optimization for Sample-Efficient Reinforcement Learning
arXiv:2601.06795v4 Announce Type: replace Abstract: Automated Theorem Proving (ATP) is a fundamental AI challenge, requiring machine-verifiable proofs in formal languages like Lean to evaluate reasoning. Reinforcement Le…
- arXiv cs.AIresearchStochastic CHAOS: Why Deterministic Inference Kills, and Distributional Variability Is the Heartbeat of Artifical Cognition
arXiv:2601.07239v2 Announce Type: replace Abstract: Deterministic inference is a comforting ideal in classical software: the same program on the same input should always produce the same output. As large language models…
- arXiv cs.AIresearchSimple Baselines are Competitive with Code Evolution
arXiv:2602.16805v2 Announce Type: replace Abstract: Code evolution is a family of techniques that rely on large language models to search through possible computer programs by evolving existing code. While code evolution…
- arXiv cs.AIresearchThe Token Games: Evaluating Language Model Reasoning with Puzzle Duels
arXiv:2602.17831v3 Announce Type: replace Abstract: Evaluating the reasoning capabilities of Large Language Models is increasingly challenging as models improve. Human curation of hard questions is highly expensive, espe…
- arXiv cs.AIresearchStrategy Executability in Mathematical Reasoning: Leveraging Human-Model Differences for Effective Guidance
arXiv:2602.22583v2 Announce Type: replace Abstract: Example-based guidance is widely used to improve mathematical reasoning at inference time, yet its effectiveness is highly unstable across problems and models-even when…
- arXiv cs.AIresearchCausal Identification from Counterfactual Data: Completeness and Bounding Results
arXiv:2602.23541v3 Announce Type: replace Abstract: Previous work establishing completeness results for counterfactual identification has been circumscribed to the setting where the input data belongs to observational or…
- arXiv cs.AIresearchDon't Ask: Understanding What Numeric Self-Reports Measure in Language Models
arXiv:2603.18893v3 Announce Type: replace Abstract: Language models are often asked to rate their internal states with a number, but it is unclear how reliable this measure is. We train probes for four affective concepts…
- arXiv cs.AIresearchScoringBench: A Benchmark for Evaluating Tabular Foundation Models with Proper Scoring Rules
arXiv:2603.29928v4 Announce Type: replace Abstract: Tabular foundation models such as TabPFN, TabICL, ExaOneTabular, and Nori already produce full predictive distributions, yet prevailing regression benchmarks evaluate t…
- arXiv cs.AIresearchMitigating Distribution Sharpening in Math RLVR via Distribution-Aligned Hint Synthesis and Backward Hint Annealing
arXiv:2604.07747v2 Announce Type: replace Abstract: In reinforcement learning with verifiable rewards (RLVR), challenging math questions whose sampled groups remain uniformly incorrect yield no informative update, lettin…
- arXiv cs.AIresearchOracleProto: A Reproducible Framework for Benchmarking LLM Forecasting via Knowledge Cutoffs and Temporal Masking
arXiv:2605.03762v2 Announce Type: replace Abstract: Forecasting is a composite capability encompassing information gathering, evidence integration, situational assessment, and decision-making. Large language models (LLMs…
- arXiv cs.AIresearchData-driven Circuit Discovery for Interpretability of Language Models
arXiv:2605.09129v2 Announce Type: replace Abstract: Circuit study aims to explain how language models (LMs) implement a task by localizing and interpreting a circuit: a computational subgraph responsible for a specific m…
- arXiv cs.AIresearchConMoE: MoE Compression as Expert Selection and Functional Assignment
arXiv:2605.29350v2 Announce Type: replace Abstract: Sparse Mixture-of-Experts (MoE) models scale model capacity through sparse activation, but storing their full parameters remains costly in GPU memory. Existing expert c…
- arXiv cs.AIresearchConformal Certification of Reasoning Trace Prefixes
arXiv:2605.30085v2 Announce Type: replace Abstract: Language model reasoning traces are rarely all-or-nothing; they frequently contain valid intermediate steps before a critical error occurs. Existing uncertainty quantif…
- arXiv cs.AIresearchRevisiting Ripple Effects in Knowledge Editing through Pressure-Aware Joint Neighborhood Optimization
arXiv:2606.01610v2 Announce Type: replace Abstract: Single-edit updates in large language models can trigger ripple effects across local knowledge neighborhoods: desirable propagation to related facts and unintended pert…
- arXiv cs.AIresearchTrajectory-Refined Distillation
arXiv:2606.08432v2 Announce Type: replace Abstract: On-policy distillation (OPD) has become a central post-training tool for large language models (LLMs), providing dense per-token teacher supervision along the student's…
- arXiv cs.AIresearchLATTEArena: An Evaluation Framework for LLM-powered Tabular Feature Engineering (Extended Version)
arXiv:2606.09004v3 Announce Type: replace Abstract: Feature engineering remains a cornerstone of tabular data analysis, and Large Language Models (LLMs) have emerged as a promising paradigm for its automation, giving ris…
- arXiv cs.AIresearchIncumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
arXiv:2606.17443v3 Announce Type: replace Abstract: Large language models (LLMs) are becoming a major way for consumers to find products, but we do not yet understand how brands compete in this new channel. We study bran…
- arXiv cs.AIresearchFrom Brewing to Resolution: Tracing the Internal Lifecycle of Code Reasoning in LLMs
arXiv:2606.17648v3 Announce Type: replace Abstract: Standard accuracy metrics cannot explain why LLMs handle variable tracking but fail on semantically equivalent loops. We study an internal lifecycle of code reasoning i…
- arXiv cs.AIresearchBeyond Trajectory Imitation: Strategy-Guided Policy Optimization for LLM Reasoning
arXiv:2606.24064v3 Announce Type: replace Abstract: Distilling reasoning capabilities from strong to weak language models typically involves imitating specific solution trajectories, effectively transferring what to answ…
- arXiv cs.AIresearchWhat a Pooled Leaderboard Does Not Say: Auditing the Scope and Provenance of Method Comparisons in Microservice RCA Benchmarks
arXiv:2606.29159v2 Announce Type: replace Abstract: Offline benchmarks for microservice root cause analysis (RCA) rank methods by a score pooled over several applications, and a pooled lead leaves two things unstated: on…
- arXiv cs.AIresearchProcess Advantage Signal Shaping: Advantage Construction for Process-Supervised Reinforcement Learning in LLM Reasoners
arXiv:2606.29296v2 Announce Type: replace Abstract: Dense process supervision supplies local feedback but leaves open how that feedback should become a policy advantage. We introduce PASS (Process Advantage Signal Shapin…