AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchLayerwise Repacking: Variable-Length Hybrid-Attention Training without an Extra Redistribution Pass
arXiv:2610.35989v1 Announce Type: cross Abstract: Variable-length training creates conflicting parallelism requirements across the layers of a hybrid-attention model. The longest samples require wide sequence-parallel (S…
- arXiv cs.AIresearchPricing Cross-Bank Context in Federated Anti-Money Laundering
arXiv:2610.35996v1 Announce Type: cross Abstract: Most graph-based Anti-Money Laundering (AML) methods assume access to a pooled transaction graph containing all accounts and payments, which in reality does not exist. Fe…
- arXiv cs.AIresearchMedJev: Evaluating Zero-Shot System-One Decision Making in Medicine
arXiv:2610.36006v1 Announce Type: cross Abstract: System One decision models such as Jev answer structured questions in a single forward pass without generating text, offering fast and computationally efficient clinical…
- arXiv cs.AIresearchDuplexTune: Coupled-State Replay for Memory-Efficient Mixture-of-Experts Adaptation
arXiv:2610.36009v1 Announce Type: cross Abstract: Sparse expert execution does not remove the stored states needed to adapt a mixture-of-experts language model. We present DuplexTune, a framework for replaying coupled hi…
- arXiv cs.AIresearchRollback: Experience Backtracking in World Models for Policy Improvement
arXiv:2610.36068v1 Announce Type: cross Abstract: Action-conditioned world models offer a scalable alternative to costly real-world interaction for improving robot policies. However, existing world-model-based methods la…
- arXiv cs.AIresearchHow Much Does the Hint Matter? Scope Conditions for Hindsight Self-Distillation in Financial Reasoning
arXiv:2610.36082v1 Announce Type: cross Abstract: Financial institutions operating under data-residency constraints are restricted to open-weight models small enough to locally host, which trail frontier systems on multi…
- arXiv cs.AIresearchDirichlet Process Mixtures within Semiparametric Bayesian Networks: A Novel Conditional Density Estimation Approach
arXiv:2610.36110v1 Announce Type: cross Abstract: Semiparametric Bayesian networks (SPBNs) are flexible probabilistic graphical models for representing joint distributions of continuous variables through lower-dimensiona…
- arXiv cs.AIresearchScaling and Renormalization in Neural Networks: An Effective Renormalization-Group Description of Grokking
arXiv:2610.36113v1 Announce Type: cross Abstract: Grokking is a delayed transition from memorization to generalization in deep neural networks (NNs). We examine whether scaling and renormalization group (RG) theories can…
- arXiv cs.AIresearchPhysical structure as an organizing principle for universal Hamiltonian learning
arXiv:2610.36135v1 Announce Type: cross Abstract: Density functional theory (DFT) gives the electronic structure of materials, but its cost limits large-supercell calculations and materials screening. Machine-learned Ham…
- arXiv cs.AIresearchBeyond the Estimate: Wiener-Informed Decoding for Easier Training of Pixel-Space Flow Models
arXiv:2610.36192v1 Announce Type: cross Abstract: Learning velocity fields directly in pixel space is difficult and slow. We show that coarse second-order image statistics can ease this task: frozen Wiener estimates impr…
- arXiv cs.AIresearchcount-catcher: Lean Theorems from Model Counting Certificates
arXiv:2610.36196v1 Announce Type: cross Abstract: We present count-catcher, a tool that turns a CPOG certificate of a count by the exact model counter d4 into a Lean theorem that a CNF formula defined in Lean has exactly…
- arXiv cs.AIresearchBase Models Can Detect Hidden Intentions that Prompted Classifiers Miss
arXiv:2610.36202v1 Announce Type: cross Abstract: The Persona Selection Model (PSM) posits that post-training selects a persona the model learned during pre-training. Work on subliminal learning has shown that models can…
- arXiv cs.AIresearchConcorde: Geometry-Aware Link Prediction via Decoupled Energy Minimization
arXiv:2610.36204v1 Announce Type: cross Abstract: Link prediction remains difficult in sparse graphs, where diffusion heuristics lack connectivity and geometric methods suffer spectral instability when coupling learned r…
- arXiv cs.AIresearchCounterfactual Profile Perturbations for Recommendation Explanations: A Reproducibility Study
arXiv:2610.36208v1 Announce Type: cross Abstract: This reproducibility paper explores the counterfactual explanation technique known as Learning to eXplain Recommendations (LXR). Prior work has shown that LXR can reliabl…
- arXiv cs.AIresearchStructure, Not Sequence: Tracking and Validating Folding Mechanisms in ESMFold's Trunk with Sparse Autoencoders
arXiv:2610.36226v1 Announce Type: cross Abstract: Structure predictors like ESMFold fold proteins using neural network computation and have transformed structural biology. However, the driving mechanism from amino-acid (…
- arXiv cs.AIresearchDirectional Statistics on Circles and Spheres: A New Angle on Data
arXiv:2610.36227v1 Announce Type: cross Abstract: Two compass bearings near north can have an ordinary arithmetic mean pointing south: the difficulty is that angles live on a circle, not a straight line. Using real wind…
- arXiv cs.AIresearchFreeze, Don't Clone: Fine-Tuning with Expressive Policies Without Retaining Offline Data
arXiv:2610.36247v1 Announce Type: cross Abstract: Offline-to-online reinforcement learning uses offline pretraining to accelerate subsequent online learning. While many methods assume the offline dataset remains availabl…
- arXiv cs.AIresearchcuArms: GPU-Accelerated Conflict-Based Planning for Multi-Arm Manipulation
arXiv:2610.36250v1 Announce Type: cross Abstract: Multi-arm manipulation has the potential to dramatically improve the throughput, efficiency, and dexterity of many manipulation tasks. Recent work coordinates multiple ro…
- arXiv cs.AIresearchDetecting and Mitigating Query-Based System Prompt Leakage Attacks
arXiv:2610.36295v1 Announce Type: cross Abstract: Large Language Models (LLMs) rely on proprietary system prompts to enforce task directives, safety policies, and access controls. These prompts are vulnerable to query-ba…
- arXiv cs.AIresearchCrowdCC: Jointly Benchmarking Corruption Robustness and Deployment Quantization in Crowd Counting
arXiv:2610.36340v2 Announce Type: replace-cross Abstract: Crowd-counting models deployed on edge devices face two challenges: inputs are degraded along the capture and transmission pipeline, while models are often quanti…
- arXiv cs.AIresearchEnhancing Clinical Classifier Robustness with Deep Representation Learning Based OOD Detection on DNA Methylation Data
arXiv:2610.36344v1 Announce Type: cross Abstract: DNA methylation (DNAM) array-based tumor classification is a powerful clinical diagnostic tool, but models trained on fixed reference cohorts can produce faulty predictio…
- arXiv cs.AIresearchReliability-Aware Short-Term Glycemic Risk Prediction Using an Anchor-Based Comparative Framework of Sequence, Multimodal, and State-Updating Digital-Twin Models
arXiv:2610.36363v1 Announce Type: cross Abstract: Anticipating short-term hypoglycemic and hyperglycemic risk is essential for proactive diabetes management. However, existing machine learning approaches often use incons…
- arXiv cs.AIresearchSOMA: A Benchmark for Fine-Grained Egocentric Behavior and Body-Motion Understanding
arXiv:2610.36384v1 Announce Type: cross Abstract: Perceiving human body motion is critical for multimodal systems that understand and assist people. Yet current evaluations of Multimodal Large Language Models (MLLMs) lar…
- arXiv cs.AIresearchExpert lithological labels do not improve learning when used to predict marine sediment properties
arXiv:2610.36410v2 Announce Type: replace-cross Abstract: Scientific drilling has produced large archives of physical, chemical and magnetic measurements linked to descriptions of the rock recovered. These archives are m…
- arXiv cs.AIresearchRepulsive Population Policy Gradients for Behaviorally Diverse Reinforcement Learning
arXiv:2610.36425v1 Announce Type: cross Abstract: The goal of Quality-Diversity Reinforcement Learning (QD-RL) is to find a population of high-performing policies that solve a decision-making problem in diverse ways. Des…
- arXiv cs.AIresearchLociArt: A Unified Paradigm for Free-Style Local Video Stylization
arXiv:2610.36456v1 Announce Type: cross Abstract: Video stylization typically transforms entire scenes, while local artistic editing requires precise region--style association, spatial control, and temporal consistency.…
- arXiv cs.AIresearchFull Collapse Is Suboptimal: Controlled Partial Neural Collapse for Long-Tailed Recognition
arXiv:2610.36460v1 Announce Type: cross Abstract: Recent long-tailed recognition (LTR) methods incorporate neural collapse (NC) and align features to a fixed Equiangular Tight Frame (ETF) as a uniform target for all clas…
- arXiv cs.AIresearchProposing Better Rollouts for On-Policy Distillation
arXiv:2610.36477v2 Announce Type: replace-cross Abstract: On-policy distillation (OPD) improves language-model post-training by letting the student learn from its own sampled trajectories under teacher supervision, align…
- arXiv cs.AIresearchProCerDCL: Query-Efficient Black-Box Probabilistic Verification of Deep Learning Compiler Optimization
arXiv:2610.36499v1 Announce Type: cross Abstract: We present \tool, a query-efficient framework for black-box probabilistic verification. Given a fixed, sampleable input distribution and a binary property oracle, \tool a…
- arXiv cs.AIresearchDo AI Scientists Know What's New?
arXiv:2610.36511v1 Announce Type: cross Abstract: A major challenge facing AI scientific discovery pipelines is how to determine when a generated idea is meaningfully novel. Ideally, novelty checks in these pipelines mus…
- arXiv cs.AIresearchFrameForge: Harnessing Priors of Video Foundation Models for Unified Image Generation
arXiv:2610.36545v1 Announce Type: cross Abstract: Unified conditional image generation requires a single model to preserve different aspects of a visual input while producing a target image that follows textual guidance.…
- arXiv cs.AIresearchREACH: An RL-based Hierarchical Humanoid Crawling Framework for Adaptable Confined Space Traversal
arXiv:2610.36546v1 Announce Type: cross Abstract: Humanoid locomotion in confined spaces requires continuous height adaptation across multiple locomotion modes, yet most existing approaches remain limited to height-adjus…
- arXiv cs.AIresearchLess is more: learning what to profile for GPU kernel optimization
arXiv:2610.36576v1 Announce Type: cross Abstract: Large language models have recently shown strong potential for automatically optimizing GPU kernels, yet their effectiveness depends critically on the hardware informatio…
- arXiv cs.AIresearchMM-Net: A Multimodal Multi-Task Network for Sleep Staging and Sleep Apnea Assessment After Ischemic Stroke
arXiv:2610.36583v1 Announce Type: cross Abstract: Automated polysomnography after ischemic stroke is challenging because of altered sleep-related neural activity and concurrent respiratory disturbances. Existing stroke-s…
- arXiv cs.AIresearchBalanced State Subspaces for Mamba-2 State Compression
arXiv:2610.36617v1 Announce Type: cross Abstract: Mamba-2 compresses all context into a fixed-size recurrent state that dominates decoding memory and bandwidth. Existing compressors quantize state values or prune state c…
- arXiv cs.AIresearchDream-COP: Guiding Algorithm Discovery for Combinatorial Optimization Problems in an Experience World
arXiv:2610.36618v1 Announce Type: cross Abstract: Large language models (LLMs) enable algorithm discovery for combinatorial optimization problems (COPs) through iterative program generation and evaluation. A central chal…
- arXiv cs.AIresearchAdapting Video Models for Robot Control via Motion-guided Asymmetric Prediction
arXiv:2610.36624v1 Announce Type: cross Abstract: World Action Models (WAMs) leverage pretrained video generative models for robotic control by jointly predicting future visual observations and actions. The visual predic…
- arXiv cs.AIresearchTap2Listen: Instant Target Switching for Streaming Audio-Visual Speaker Extraction
arXiv:2610.36643v1 Announce Type: cross Abstract: Streaming audio-visual target speaker extraction (AV-TSE) uses a target speaker's face to extract the speaker's speech from an input audio mixture in real time. However,…
- arXiv cs.AIresearchA Quantitative Approach to Asch Conformity in Frontier Large Language Models
arXiv:2610.36668v1 Announce Type: cross Abstract: The Asch conformity experiment has recently gained considerable attention as a way of measuring conformity in large language models (LLMs). While previous studies have fo…
- arXiv cs.AIresearchLook Only When Needed: Certificate-Guided Visual Routing for Efficient Diffusion VLMs
arXiv:2610.36671v1 Announce Type: cross Abstract: Diffusion vision-language models (DVLMs) process their entire input at every denoising step. Since visual tokens make up 89-97% of this input on the benchmarks we study,…
- arXiv cs.AIresearchSpecialization, Diversity, and Scaling in Child-Scale Language Models
arXiv:2610.36679v1 Announce Type: cross Abstract: Modern language models are trained on far more language than children experience, but we know much less about learning from small, naturalistic datasets. We train small G…
- arXiv cs.AIresearchTOPS: Reshaping Flow Source and Transport for Multi-Task Robot Manipulation
arXiv:2610.36694v1 Announce Type: cross Abstract: Flow-based policies for language-conditioned multi-task manipulation commonly share a single $\textit{Gaussian}$ source prior and a common linear transport path across ta…
- arXiv cs.AIresearchHeteGCD: Statistics-Aware Dual-Prototype Learning for Generalized Category Discovery on Heterophilous Graphs
arXiv:2610.36696v1 Announce Type: cross Abstract: Heterophilous graphs are prevalent in real-world applications, where connected nodes often exhibit dissimilar attributes or belong to different classes, posing significan…
- arXiv cs.AIresearchConservative Dueling Bandits: Preference Learning under Baseline Constraints
arXiv:2610.36708v1 Announce Type: cross Abstract: Contextual dueling bandits learn from pairwise comparisons, the same feedback used to fine-tune language models from human preferences. Standard algorithms explore aggres…
- arXiv cs.AIresearchGenerative Refinement Learning for Temporal Interpolation in Scientific Machine Learning
arXiv:2610.36751v1 Announce Type: cross Abstract: Computationally expensive simulations and real-world observations often provide data at a lower temporal resolution than downstream tasks require. Averaging observations…
- arXiv cs.AIresearchRobust LLM Data Selection via Content-Invariant Gradient Attribution
arXiv:2610.36755v1 Announce Type: cross Abstract: Gradient-based data attribution has become an important tool for high-quality data selection in the instruction fine-tuning of large language models (LLMs), as it estimat…
- arXiv cs.AIresearchMaking Self-Driving Laboratories Auditable:A Unified Framework for Hybrid Actions,Measurements, and Safe Execution
arXiv:2610.36761v1 Announce Type: cross Abstract: Self-driving laboratories (SDLs) are laboratory systems that couple experimental planning, automated execution, and measurement feedback in a closed loop. Integrating fle…
- arXiv cs.AIresearchThe Vocabulary Is Already Here: Adapting Pretrained Interpretable Features for Generative Recommendation
arXiv:2610.36763v1 Announce Type: cross Abstract: Generative recommendation (GR) predicts the next item through its semantic ID (SID). Existing methods commonly fit SID vocabularies to pretrained language model (PLM) rep…
- arXiv cs.AIresearchSelect-TTT: Query-Conditioned Selective Fast Weight Updates for In-Place Test-Time Training
arXiv:2610.36794v1 Announce Type: cross Abstract: Test-time training (TTT) enables a deployed language model to internalize incoming context on-the-fly into a small set of fast-weight parameters, rather than merely treat…
- arXiv cs.AIresearchAn Algebra of Transformers
arXiv:2610.36795v1 Announce Type: cross Abstract: On the theoretic analyses on transforms, they mainly focus on the computability analysis, such the Turing-completeness of unbounded precision transformers, the advantage…