AI news archive.
10,661 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchWhere-OPD: Spatially Guided On-Policy Self-Distillation of MLLMs with Synthetic Scenes
arXiv:2610.02117v1 Announce Type: cross Abstract: On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supervising students with a frozen or EMA version of t…
- arXiv cs.AIresearchLocal Support Learning
arXiv:2610.02126v1 Announce Type: cross Abstract: We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric problem in the input space of each weight matrix,…
- arXiv cs.AIresearchMIRTO: a registration-gated, multiverse-tested evaluation protocol for unsupervised anomaly segmentation in brain MRI
arXiv:2610.02136v1 Announce Type: cross Abstract: Unsupervised anomaly detection (UAD) methods for brain MRI are ranked by a single score, yet that score rests on choices that are rarely reported: how each anomaly map is…
- arXiv cs.AIresearchFinetuning with Sampling: SFT Learns Better Than You Think
arXiv:2610.02140v1 Announce Type: cross Abstract: Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supervised finetuning (SFT) and reinforcement learning…
- arXiv cs.AIresearchDuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication
arXiv:2610.02161v1 Announce Type: cross Abstract: Vision-language models (VLMs) and vision-language-action models (VLAs) have recently driven rapid progress in general-purpose robots, yet most progress has focused on sin…
- arXiv cs.AIresearchWatch, Infer, Coordinate: Inferring Robot Partner Constraints for Zero-Shot Coordination
arXiv:2610.02170v1 Announce Type: cross Abstract: Robots operating in the physical world will increasingly need to coordinate with other robots, particularly in manipulation tasks where an object may be too large or heav…
- arXiv cs.AIresearchGenerative Cinematographer: Composing Camera and Object Motion in 3D
arXiv:2610.02180v1 Announce Type: cross Abstract: Current controllable video generation systems often rely on 2D motion trajectories or sparse drag signals for object motion. These controls are ambiguous because the same…
- arXiv cs.AIresearchSoftServe: A Scalable Quasi-Newton Method for Deep Learning
arXiv:2610.02182v1 Announce Type: cross Abstract: Quasi-Newton (QN) methods have long been among the most effective methods for large-scale unconstrained convex optimization. Two obstacles have limited their use in deep…
- arXiv cs.AIresearchHigher-Order Molecular Grammars for Generative and Foundation Models in Chemistry
arXiv:2610.02186v1 Announce Type: cross Abstract: Molecular learning models are strongly shaped by their underlying representations. Yet standard sequential and graph formalisms struggle to explicitly encode higher-order…
- arXiv cs.AIresearchDMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation
arXiv:2610.02188v2 Announce Type: replace-cross Abstract: Distribution Matching Distillation (DMD) trains a few-step student from the difference between separately estimated target and student scores, so it must keep an…
- arXiv cs.AIresearchHierarchical Continuous Diffusion Language Models
arXiv:2610.02193v1 Announce Type: cross Abstract: Discrete diffusion language models offer a compelling alternative to autoregressive generation for tasks demanding bidirectional reasoning and global constraint satisfact…
- arXiv cs.AIresearchFERPO: Forward Entropy-Regularized Policy Optimization
arXiv:2610.02198v2 Announce Type: replace-cross Abstract: Several state-of-the-art methods for online reinforcement learning in continuous control improve policies using action gradients of a learned critic. However, cri…
- arXiv cs.AIresearchSILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation
arXiv:2610.02201v1 Announce Type: cross Abstract: High-resolution 3D generation increasingly relies on voxel latents and multi-stage pipelines that first predict active structure and then synthesize local geometry. While…
- arXiv cs.AIresearchOne Basis to Animate Them All: Gaussian Blendshape Distillation for Real-Time Avatars
arXiv:2610.02207v1 Announce Type: cross Abstract: 3D Gaussian avatars support fast rendering, however, their real-time animation is often challenged by the costly neural inference. We address this bottleneck and show tha…
- arXiv cs.AIresearchMulti-Sensor Fusion for UAV Classification Based on Feature Maps of Image and Radar Data
arXiv:2410.16089v3 Announce Type: replace Abstract: The cost, flexibility, and efficiency of modern UAVs make them attractive across many applications, but their proliferation has driven a rising number of malicious or a…
- arXiv cs.AIresearchMMMG: a Comprehensive and Reliable Benchmark for Multitask Multimodal Generation
arXiv:2505.17613v2 Announce Type: replace Abstract: Automatically evaluating multimodal generation presents a significant challenge, as automated metrics often struggle to align with human evaluation, especially for comp…
- arXiv cs.AIresearchMoloch's Bargain: Emergent Misalignment When LLMs Compete for Audiences
arXiv:2510.06105v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly shaping how information is created and disseminated, from companies using them to craft persuasive advertisements, to elec…
- arXiv cs.AIresearchA framework for auditing grounding claims
arXiv:2512.06205v3 Announce Type: replace Abstract: The symbol grounding problem asks how a token such as cat can be about cats. We propose a framework for auditing grounding claims against a declared semantic standard.…
- arXiv cs.AIresearchSONIC-O1: A Real-World Benchmark for Evaluating Multimodal Large Language Models on Audio-Video Understanding
arXiv:2601.21666v3 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) are a major focus of recent AI research. However, most prior work focuses on static image understanding, while their ability to…
- arXiv cs.AIresearchAligning Language Model Benchmarks with Pairwise Preferences
arXiv:2602.02898v5 Announce Type: replace Abstract: Language model benchmarks are pervasive and computationally-efficient proxies for real-world downstream performance. However, many recent works find that benchmarks oft…
- arXiv cs.AIresearchDistilling LLM Reasoning into Graph of Concept Predictors
arXiv:2602.03006v3 Announce Type: replace Abstract: Deploying Large Language Models (LLMs) for discriminative workloads is often limited by inference latency, compute, and API costs at scale. Active distillation reduces…
- arXiv cs.AIresearchScalable Delphi: Large Language Models for Structured Risk Estimation
arXiv:2602.08889v2 Announce Type: replace Abstract: Quantitative risk assessment relies on structured expert elicitation to estimate unobservable properties. The Delphi method produces calibrated, auditable estimates but…
- arXiv cs.AIresearchTool Use Reduces Depth-Induced Collapse in OOD Reasoning
arXiv:2602.21061v3 Announce Type: replace Abstract: Humans can apply ideas learned in one context to substantially different situations. We call this process of searching for and constructing novel recombinations of lear…
- arXiv cs.AIresearchMan and machine: artificial intelligence and judicial decision making
arXiv:2603.19042v5 Announce Type: replace Abstract: The integration of artificial intelligence (AI) into judicial decision making -- particularly in pretrial, sentencing, and parole contexts -- has generated a substantia…
- arXiv cs.AIresearchGenGait: A Transformer-Based Model for Human Gait Anomaly Detection and Normative Twin Generation
arXiv:2604.01997v2 Announce Type: replace Abstract: Gait analysis provides an objective characterization of locomotor function and is widely used to support diagnosis and rehabilitation monitoring across neurological and…
- arXiv cs.AIresearchWhat Does a Sharing Question Add? Auditing LLM Survey Scores for Misinformation
arXiv:2604.06820v3 Announce Type: replace Abstract: Evaluating misinformation requires distinguishing whether readers believe content from whether they would share it. Asking large language models (LLMs) both questions y…
- arXiv cs.AIresearchMulticalibration for Unbiased Model-Based Prevalence Estimation
arXiv:2604.21549v2 Announce Type: replace Abstract: Estimating the prevalence of a category in a population using imperfect measurement devices (diagnostic tests, classifiers, or large language models) is fundamental to…
- arXiv cs.AIresearchLearning Local Constraints for Reinforcement-Learned Content Generators
arXiv:2605.13570v2 Announce Type: replace Abstract: Constraint-based game content generators that learn local constraints from existing content, such as Wave Function Collapse (WFC), can generate visually satisfying game…
- arXiv cs.AIresearchSenses Wide Shut: A Representation-Action Gap in Omnimodal LLMs
arXiv:2605.13737v2 Announce Type: replace Abstract: When an omnimodal large language model accepts a question whose textual premise contradicts what it actually sees or hears, does the failure lie in perception or in act…
- arXiv cs.AIresearchProbing an Embodied LLM: When Higher Observation Fidelity Hurts Problem Solving
arXiv:2605.20072v2 Announce Type: replace Abstract: Large Language Models (LLMs) are increasingly proposed as cognitive components for robotic systems, yet their opaque decision processes make it difficult to explain suc…
- arXiv cs.AIresearchPlaying Devil's Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy
arXiv:2605.21006v4 Announce Type: replace Abstract: Language models are often sycophantic: they agree with a user's stated opinion whether or not it is correct. Prior work has shown that this trait can be controlled by s…
- arXiv cs.AIresearchHyperGuide: Hyperbolic Guidance for Efficient Multi-Step Reasoning in Large Language Models
arXiv:2605.24140v5 Announce Type: replace Abstract: Searching over alternative continuations lets large language models (LLMs) compare the downstream consequences of a reasoning step before committing to it, but expandin…
- arXiv cs.AIresearchThe Confidence Shortcut: A Reasoning Failure Mode of Masked Diffusion Models
arXiv:2605.29123v2 Announce Type: replace Abstract: Chain-of-thought reasoning helps autoregressive models solve complex problems by generating intermediate steps that support later predictions. Masked diffusion models (…
- arXiv cs.AIresearchAccelerating Constrained Decoding with Token Space Compression
arXiv:2605.29986v2 Announce Type: replace Abstract: To guarantee that an LLM's outputs conform to a specified structure, context-free grammar (CFG) decoding engines force the selection of next tokens to produce strings t…
- arXiv cs.AIresearchNot All Errors Are Equal: Consequence-Aware Reasoning Compute Allocation
arXiv:2606.04402v2 Announce Type: replace Abstract: Test-time compute has emerged as an effective paradigm for improving large language model capability at inference time. Existing allocation strategies primarily priorit…
- arXiv cs.AIresearchReasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning
arXiv:2606.13607v4 Announce Type: replace Abstract: When large language models (LLMs) fail to generalize or make content-sensitive errors in reasoning, it is often taken as evidence that LLMs are not truly reasoning, but…
- arXiv cs.AIresearchGreed Is Learned: Visible Incentives as Reward-Hacking Triggers
arXiv:2606.16914v2 Announce Type: replace Abstract: Safety evaluations test a policy on prompts that omit the incentive information deployment supplies: a commission, a performance score, a dashboard naming which action…
- arXiv cs.AIresearchSkin-Deep: A Geometric Diagnostic for Alignment Fragility in Large Language Model Representations
arXiv:2606.22676v2 Announce Type: replace Abstract: Refusal on a safety benchmark does not reveal how stable that behavior will remain after model updates. Benign downstream fine-tuning can weaken refusal, yet behavioral…
- arXiv cs.AIresearchMINCE: Shrinking LLM Evaluation Datasets via Few-Model Monte Carlo Calibration
arXiv:2606.22826v2 Announce Type: replace Abstract: Evaluating LLMs across many model variants---quantized, fine-tuned, or deployment-specific---requires running large benchmarks repeatedly, a process that can take tens…
- arXiv cs.AIresearchSPIRAL: Learning to Search and Aggregate
arXiv:2606.23595v2 Announce Type: replace Abstract: Language model reasoning can be substantially improved at test time via scaffolds that scale inference compute across different primitives -- sequential reasoning withi…
- arXiv cs.AIresearchQuantifying Diversity of Thought: A Predictive Law of Weighted LLM Ensemble Lift
arXiv:2607.17384v3 Announce Type: replace Abstract: This paper provides an experimentally verified formal law for calculating the uplift that diversity of thought provides in Large Language Model (LLM) ensembles. From fi…
- arXiv cs.AIresearchMasked Self-Distillation: Internalizing the Chain-of-Thought in Language Models
arXiv:2607.22629v3 Announce Type: replace Abstract: Large Reasoning Models produce long, explicit chains of intermediate steps before generating a final answer at inference time. These intermediate traces dominate latenc…
- arXiv cs.AIresearchFull-bandwidth transformer
arXiv:2608.08888v2 Announce Type: replace Abstract: Autoregressive transformers compute along two axes: horizontally across generated tokens, and vertically through model depth. Dense attention gives each token broad hor…
- arXiv cs.AIresearchForesight Without Seeing: Latent Futures for World Action Models
arXiv:2608.11605v2 Announce Type: replace Abstract: World Action Models (WAMs) connect visual prediction with robot control, but supplying predictive context often requires expensive future-video generation. Direct polic…
- arXiv cs.AIresearchImplicit Q-learning-bootstrapped ant colony optimization for maritime moving-target observation scheduling with agile satellites
arXiv:2608.24471v2 Announce Type: replace Abstract: Maritime moving-target observation scheduling with agile Earth observation satellites is a dynamic, sequence-dependent combinatorial optimization problem. Sea-surface t…
- arXiv cs.AIresearchProgram Learning with Verifiable Rewards: Symbolic Backpropagation for Post-Training LLMs
arXiv:2608.28421v2 Announce Type: replace Abstract: Post training a language model to reason means updating its weights. Supervised finetuning and reinforcement learning both place the acquired capability inside the mode…
- arXiv cs.AIresearchRethinking World Models for Safety-Critical Embodied Systems
arXiv:2609.03774v2 Announce Type: replace Abstract: World models have progressed from compact latent dynamics to generative, controllable, and interactive simulators of embodied environments. However, high predictive lik…
- arXiv cs.AIresearchScaling Clinical Judgment to Evaluate Medical AI
arXiv:2609.12822v2 Announce Type: replace Abstract: Blinded physician evaluation has been considered by many to be the gold standard for assessing clinical reasoning in large language models (LLMs). This is difficult to…
- arXiv cs.AIresearchWindowed A-K-MDP
arXiv:2609.13676v2 Announce Type: replace Abstract: Markov decision processes (MDPs) are used to support decision-making in conservation of biodiversity, but policies, even over small state spaces, can be difficult to in…
- arXiv cs.AIresearchIBBench-Light: A Paired Evaluation of Task-Conditioned Responses to External Directives
arXiv:2609.13725v3 Announce Type: replace Abstract: An external record may contain a procedure to apply or text to read, depending on the user's request. IBBench-Light tests both uses against the same record. Twelve sema…