AI news archive.
10,660 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchViTeX-Bench: Benchmarking High-Fidelity Video Scene Text Editing
arXiv:2609.40356v2 Announce Type: replace-cross Abstract: Recent video generation is increasingly realistic and controllable, yet video editing remains less developed, particularly for precise local edits that must prese…
- arXiv cs.AIresearchSemifactual Credit-Augmented Policy Optimization
arXiv:2609.40360v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has improved the reasoning capabilities of large language models (LLMs), yet their predictions remain sensitive to t…
- arXiv cs.AIresearchGeneralising from Self-Produced Data: Model Training Beyond Human Constraints
arXiv:2504.04711v3 Announce Type: replace Abstract: Current large language models (LLMs) are constrained by human-derived training data and limited by a single level of abstraction that impedes definitive truth judgments…
- arXiv cs.AIresearchHierarchical Reasoning Model
arXiv:2506.21734v4 Announce Type: replace Abstract: Reasoning, the process of devising and executing complex goal-oriented action sequences, remains a critical challenge in AI. Current large language models (LLMs) primar…
- arXiv cs.AIresearchA Quantitative Study of Sustained Focus in Large Language Models via Repetitive Deterministic Prediction Tasks
arXiv:2511.00763v3 Announce Type: replace Abstract: We investigate the performance of large language models (LLMs) on repetitive deterministic prediction tasks and study how the sequence accuracy rate (SAR) scales with o…
- arXiv cs.AIresearchSuperficial Reflection or Genuine Thought? A Fine-Grained Cognitive Analysis of Large Reasoning Models
arXiv:2512.00729v2 Announce Type: replace Abstract: Motivated by the observed human-like behaviours in Large Reasoning Models (LRMs), this paper introduces a comprehensive taxonomy to characterise atomic reasoning steps…
- arXiv cs.AIresearchAn Empirical Study of On-Device Translation for Real-Time Live-Stream Chat on Mobile Devices
arXiv:2601.02641v3 Announce Type: replace Abstract: Despite its efficiency, there has been little research on the practical aspects required for real-world deployment of on-device AI models, such as the device's CPU util…
- arXiv cs.AIresearchSurvival is the Only Reward: Sustainable Self-Training Through Environment-Mediated Selection
arXiv:2601.12310v3 Announce Type: replace Abstract: Self-training systems often degenerate due to the lack of an external criterion for judging data quality, leading to reward hacking and semantic drift. This paper provi…
- arXiv cs.AIresearchLSR-Ben: A Logical and Scientific Reasoning Benchmark for Evaluating Process Reward Models
arXiv:2605.01203v3 Announce Type: replace Abstract: Currently, process reward models (PRMs) have exhibited remarkable potential for test-time scaling. Since large language models (LLMs) regularly generate flawed intermed…
- arXiv cs.AIresearchSTEPS: Selective On-Policy Self-Distillation for Reasoning
arXiv:2605.10194v2 Announce Type: replace Abstract: On-policy self-distillation (OPSD) uses a model as its own teacher under privileged context, providing token-level supervision on the model's own reasoning trajectories…
- arXiv cs.AIresearch$D^2$-Monitor: Dynamic Safety Monitoring for Diffusion LLMs via Hesitation Signals
arXiv:2605.25893v2 Announce Type: replace Abstract: Despite the emergence of diffusion large language models (D-LLMs) as an alternative to autoregressive large language models (AR-LLMs), safety monitoring for D-LLMs rema…
- arXiv cs.AIresearchSTAB: Specification-driven Testing for Algorithmic Bottlenecks
arXiv:2605.27981v2 Announce Type: replace Abstract: Evaluating the efficiency of algorithmic code requires test cases that expose runtime bottlenecks. Previous methods generate efficiency test cases either by increasing…
- arXiv cs.AIresearchEmergent alignment and the projectability of ethical personas
arXiv:2606.09475v3 Announce Type: replace Abstract: Work on `emergent misalignment' shows that finetuning LLMs on narrow tasks can induce broadly misaligned behavior. This supports the `persona selection' (PSM) hypothesi…
- arXiv cs.AIresearchCombEval: A Framework for Evaluating Combinatorial Counting in Large Language Models
arXiv:2606.19788v2 Announce Type: replace Abstract: We present CombEval, a dynamic benchmark for evaluating combinatorial counting in large language models. CombEval represents each problem as a typed Cofola specificatio…
- arXiv cs.AIresearchGroup Preference Collapse in Personalized Multimodal Large Language Models
arXiv:2607.22603v2 Announce Type: replace Abstract: Personalized multimodal large language models (MLLMs) aim to generate user-specific responses, but existing methods mainly rely on profile-level information and overloo…
- arXiv cs.AIresearchBeyond Rephrasing: Book-Level Organization Improves Synthetic Textbook Data for Mid-Training
arXiv:2607.28109v2 Announce Type: replace Abstract: Synthetic textbook data has improved language model pre-training, but prior work largely treats the benefit as a property of generated content or local rewriting style.…
- arXiv cs.AIresearchFast LapSum: Exact Differentiable Top-$k$ at Million Scale
arXiv:2608.06912v3 Announce Type: replace Abstract: Selecting the top-$k$ elements is a fundamental operation for inducing sparsity in large-scale models and optimization problems, enabling robust expert activation, toke…
- arXiv cs.AIresearchWho Bridges Safety? Identifying and Targeting Cross-Lingual Shared Safety Pathways
arXiv:2608.09095v2 Announce Type: replace Abstract: Uncovering the internal mechanisms underlying the safety capabilities of large language models (LLMs) is crucial for developing trustworthy artificial intelligence. Cur…
- arXiv cs.AIresearchMAE I Trust Myself? Self-Evaluating VLA Action Generation with Markov Attention Entropy
arXiv:2608.16697v2 Announce Type: replace Abstract: Vision-Language-Action models (VLAs) integrate visual perception, language instruction, and action generation into end-to-end policies across heterogeneous architecture…
- arXiv cs.AIresearchLearning What to Practice: Diagnosis-Guided Self-Evolution for Language Models
arXiv:2609.00768v2 Announce Type: replace Abstract: Self-play supports the self-evolution of language models, but solver performance can plateau or decline across rounds without guidance. Existing unguided methods typica…
- arXiv cs.AIresearchMolecular D\'ej\`a Vu: Digit-Level Retrieval of Molecular Properties in Frontier Language Models
arXiv:2609.05381v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly employed to predict molecular properties. However, prediction error alone cannot distinguish prediction from retrieval of…
- arXiv cs.AIresearchA polynomial time algebraic solution to exact marginal inference in Markov Random Field models
arXiv:1709.09051v3 Announce Type: replace-cross Abstract: This paper develops on algebraic grounds a polynomial time exact linear solution to the hard combinatorial problem of marginal inference in Markov random field (M…
- arXiv cs.AIresearchETHER: Aligning Emergent Communication for Hindsight Experience Replay
arXiv:2307.15494v4 Announce Type: replace-cross Abstract: Hindsight Experience Replay (HER) enhances sample efficiency in goal-conditioned reinforcement learning (RL) by relabelling failed trajectories with goals that we…
- arXiv cs.AIresearchSpatio-Temporal Partial Sensing Forecast for Long-term Traffic
arXiv:2408.02689v3 Announce Type: replace-cross Abstract: Traffic forecasting uses recent measurements by sensors installed at chosen locations to forecast the future road traffic. Existing work either assumes all locati…
- arXiv cs.AIresearchMitigating Memorization In Language Models
arXiv:2410.02159v3 Announce Type: replace-cross Abstract: Language models (LMs) can "memorize" information, i.e., encode training data in their weights in such a way that inference-time queries can lead to verbatim regur…
- arXiv cs.AIresearchBringing Generative Learning to Representation Learning: Self-Supervised Transfer Learning as Distribution Matching
arXiv:2502.14424v5 Announce Type: replace-cross Abstract: Most self-supervised learning objectives defend against collapse but leave the target representation law unspecified. We formulate representation learning as Dist…
- arXiv cs.AIresearchAneumoBench: A Source-Linked Benchmark for Synthetic-Geometry Transfer in Aneurysm CFD
arXiv:2505.14717v2 Announce Type: replace-cross Abstract: Scientific machine learning uses simulation data to train surrogate models for fast physical-field prediction across geometries. Local shape editing can expand li…
- arXiv cs.AIresearchCross-Layer Discrete Concept Discovery for Interpreting Language Models
arXiv:2506.20040v4 Announce Type: replace-cross Abstract: Interpreting language models remains challenging due to the existence of residual stream, which linearly mixes and duplicates features across adjacent layers, cau…
- arXiv cs.AIresearchNMIXX: Domain-Adapted Neural Embeddings for Cross-Lingual eXploration of Finance
arXiv:2507.09601v3 Announce Type: replace-cross Abstract: Financial text embeddings must distinguish changes in event status, perspective, and obligations even when passages share similar wording. NMIXX adapts existing e…
- arXiv cs.AIresearch3D Software Synthesis Driven by Constraint-Expressive Intermediate Representation
arXiv:2507.18625v3 Announce Type: replace-cross Abstract: Graphical user interface (UI) software has undergone a fundamental transformation from traditional two-dimensional (2D) desktop/web/mobile interfaces to spatial t…
- arXiv cs.AIresearchOut-of-Distribution Detection using Counterfactual Distance
arXiv:2508.10148v2 Announce Type: replace-cross Abstract: Accurate and explainable out-of-distribution (OOD) detection is required to use machine learning systems safely. Previous work has shown that feature distance to…
- arXiv cs.AIresearchFrom Construction to Injection: Edit-Based Fingerprints for Large Language Models
arXiv:2509.03122v5 Announce Type: replace-cross Abstract: Reliable model fingerprints are essential for protecting large language models (LLMs) against unauthorized redistribution and commercial misuse. In black-box depl…
- arXiv cs.AIresearchThink Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression
arXiv:2510.01581v2 Announce Type: replace-cross Abstract: Recent thinking models are capable of solving complex reasoning tasks by scaling test-time compute, but this scaling should be allocated in line with task difficu…
- arXiv cs.AIresearchInstruction Retrieval at Inference Time for Small Language Models
arXiv:2510.13935v3 Announce Type: replace-cross Abstract: The facts a language model stores are tied to its parameter count, so small models that fit on edge devices fail on expert problems, which need specialized knowle…
- arXiv cs.AIresearchFrom Charts to Code: A Hierarchical Benchmark for Multimodal Models
arXiv:2510.17932v5 Announce Type: replace-cross Abstract: We introduce Chart2Code, a new benchmark for evaluating the chart understanding and code generation capabilities of large multimodal models (LMMs). Chart2Code is…
- arXiv cs.AIresearchMedRECT: A Bilingual Medical Reasoning Benchmark for Error Correction in Clinical Texts
arXiv:2511.00421v2 Announce Type: replace-cross Abstract: Large language models (LLMs) show promise in medical applications, but their ability to detect and correct errors in clinical texts remains under-evaluated, parti…
- arXiv cs.AIresearchFoCLIP: A Feature-Space Misalignment Framework for CLIP-Based Image Manipulation and Detection
arXiv:2511.06947v2 Announce Type: replace-cross Abstract: The well-aligned attribute of CLIP-based models enables its effective application like CLIPscore as a widely adopted image quality assessment metric. However, suc…
- arXiv cs.AIresearchEDGC: Entropy-driven Dynamic Gradient Compression for Efficient LLM Training
arXiv:2511.10333v2 Announce Type: replace-cross Abstract: Training large language models (LLMs) at scale incurs substantial communication overhead, while static gradient compression cannot adapt to gradient evolution and…
- arXiv cs.AIresearchFast Generalized Neural Tangent Kernel Statistics via Trace Estimation
arXiv:2511.10796v2 Announce Type: replace-cross Abstract: The empirical state-space Neural Tangent Kernel (NTK) describes the local learning geometry of a finite-width neural network, but computing it explicitly is almos…
- arXiv cs.AIresearchGraphToxin: Reconstructing Full Unlearned Graphs from Graph Unlearning
arXiv:2511.10936v3 Announce Type: replace-cross Abstract: Graph unlearning (GU) has emerged as a promising solution to comply with "the right to be forgotten" regulations by enabling the removal of sensitive information…
- arXiv cs.AIresearchGenerative AI Purpose-built for Social and Mental Health: A Real-World Pilot
arXiv:2511.11689v4 Announce Type: replace-cross Abstract: Generative AI chatbots built for mental health could extend access to care, but evidence from real-world use is limited. We report a single-arm, naturalistic pilo…
- arXiv cs.AIresearchFrom Compound Figures to Medical Multi-image Reasoning: Scaling Multimodal Large Language Models with Biomedical Literature
arXiv:2511.22232v2 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) are increasingly capable in medical imaging, yet most focus on single-image settings. Clinical interpretation often requi…
- arXiv cs.AIresearchProvable Benefit of SignGD: A Minimal Model Under Heavy-Tailed Class Imbalance
arXiv:2512.00763v2 Announce Type: replace-cross Abstract: Adaptive and non-Euclidean optimizers often outperform Euclidean methods such as stochastic gradient descent (SGD) in language modeling by a large margin. Existin…
- arXiv cs.AIresearchDiffusion-Augmented Markov Decision Processes for Maximum Entropy Reinforcement Learning
arXiv:2512.02019v4 Announce Type: replace-cross Abstract: Diffusion models provide an expressive framework for sampling from complex, unnormalized distributions. In this work, we extend Maximum Entropy Reinforcement Lear…
- arXiv cs.AIresearchActive Learning with Imperfect Labels: Optimal Labeler Assignment and Sample Selection
arXiv:2512.12870v2 Announce Type: replace-cross Abstract: Active Learning (AL) is commonly used in applications where labeling data is expensive or time-consuming. In practice, however, labels are often noisy due to vary…
- arXiv cs.AIresearchEdge-Aware and Content-Adaptive Infrared Gas Leak Detection for Industrial Safety Monitoring
arXiv:2512.23234v4 Announce Type: replace-cross Abstract: Infrared gas leak detection is important for industrial safety and environmental monitoring, but automatic detection remains challenging because gas plumes are of…
- arXiv cs.AIresearchContext-Aware Classification and Grading of Sensitive Information in Online Conversational Health Data
arXiv:2601.09717v2 Announce Type: replace-cross Abstract: Online medical consultations contain sensitive health information whose privacy implications depend not only on the entities mentioned but also on how those entit…
- arXiv cs.AIresearchSpace Filling Curves is All You Need: Communication-Avoiding Matrix Multiplication Made Simple
arXiv:2601.16294v3 Announce Type: replace-cross Abstract: General Matrix Multiplication (GEMM) is the cornerstone of HPC workloads and Deep Learning. State-of-the-art (SOTA) vendor libraries tune tensor layouts, parallel…
- arXiv cs.AIresearchART for Diffusion Sampling: A Reinforcement Learning Approach to Timestep Schedule
arXiv:2601.18681v3 Announce Type: replace-cross Abstract: We consider time discretization for score-based diffusion models to generate samples from a learned reverse-time dynamic on a finite grid. Uniform and hand-crafte…
- arXiv cs.AIresearchFunctional Subspace, where language models can use vector algebra to solve problems
arXiv:2602.01687v3 Announce Type: replace-cross Abstract: Large language models (LLMs) were invented for natural language tasks such as translation, but they have proved that they can perform highly complex functions acr…