AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchUnified Multiparadigm Fusion Intelligence for Rare Disease Diagnosis
arXiv:2610.24024v1 Announce Type: new Abstract: Rare diseases (RDs) affect an estimated 300 million people globally, leaving patients to navigate a protracted diagnostic odyssey constrained by clinical scarcity and fragm…
- arXiv cs.AIresearchLogicVal:A Logic-Formalized Benchmark for Constraint Satisfaction in Text-to-Image Generation
arXiv:2610.24051v1 Announce Type: new Abstract: Text-to-Image (T2I) generative models have made significant progress in producing high-fidelity images from natural-language prompts. Alongside this progress, numerous benc…
- arXiv cs.AIagentsWorld of SWECraft: Agent Collaboration at Scale via Simulated Software Worlds
arXiv:2610.24061v2 Announce Type: replace Abstract: As teams of AI agents take on larger projects and long-standing open problems, agent collaboration offers another dimension for scaling inference-time compute. To under…
- arXiv cs.AIagentsPersistent Memory and Iterative Improvement in an LLM Agent for Chinese-Character Drawing
arXiv:2610.24090v1 Announce Type: new Abstract: Does persistent memory accumulated from previous attempts help an LLM agent solve new problems, and does memory design affect that benefit? We study programmatic Chinese-ch…
- arXiv cs.AIresearchCache2LoRA: Understanding and Improving Context Internalization
arXiv:2610.24107v1 Announce Type: new Abstract: Context-to-LoRA methods encode documents as low-rank parameter updates, allowing language models to answer without retaining the source text in their decoding context. How…
- arXiv cs.AIagentsJoint, Decoupled, or Agent: When Each Multimodal Alignment Paradigm Wins
arXiv:2610.24108v1 Announce Type: new Abstract: Jointly pretrained multimodal encoders are a common way to combine modalities, while alternatives integrate their information during downstream training or inference. When…
- arXiv cs.AIresearchA Conditional No-Hair Theorem for Transformers: How Language Models Forget Their Prompt
arXiv:2610.24133v1 Announce Type: new Abstract: We ask how much an autoregressive transformer retains of its prompt after generating $t$ tokens. Consider two different prompts followed by the same continuation text, so t…
- arXiv cs.AIresearchTorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU
arXiv:2610.24161v4 Announce Type: replace Abstract: TorchTPU is an open-source collaboration between Google and Meta that introduces a native and principled PyTorch backend for Google TPUs. It enables seamless out-of-the…
- arXiv cs.AIagentsSelf-Improving Video-to-Action Models
arXiv:2610.24177v1 Announce Type: new Abstract: Video generative models are an emerging tool for embodied agents: they can serve as scalable visual planners, generating videos that depict how to complete previously unsee…
- arXiv cs.AIagentsAgents' Last Doubt: Knowing When to Act and When to Step Back in Tool-Using Agents
arXiv:2610.24194v1 Announce Type: new Abstract: Tool-using LLM agents repeatedly face decision points under uncertainty, choosing between autonomous action, additional information gathering, and human escalation. These d…
- arXiv cs.AIresearchDistillation, Self-Training, and the Fragility of Learned Chemical Reasoning
arXiv:2610.24199v1 Announce Type: new Abstract: Automating synthesis planning for small-molecule drug design requires chemical reasoners that are accurate as well as open to inspection and adaptation, yet frontier langua…
- arXiv cs.AIresearchTraining-Free Prompt Selection for Efficient Reinforcement Learning With Partial Rollouts
arXiv:2610.24229v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has emerged as a prominent paradigm for training reasoning models, but its computational cost makes efficient data sel…
- arXiv cs.AIresearchProbing Ambiguity: Adversarial Intent Resolution for Robust Text-to-SQL
arXiv:2610.24248v1 Announce Type: new Abstract: Text-to-SQL systems typically assume a one-to-one mapping between natural-language queries and correct SQL translations. In practice, however, many queries are inherently a…
- arXiv cs.AIagentsCortex: A Fixed-Point Theory of Governed Coding Agents - Technical Report
arXiv:2610.24255v1 Announce Type: new Abstract: A coding agent combines a large language model (LLM) with a harness that plans, edits, and executes code. We study Cortex, a supervisory layer for pre-execution governance,…
- arXiv cs.AIagentsAM-Bench: A Unified Taxonomy and Evaluation Suite for Agentic Misalignment
arXiv:2610.24264v3 Announce Type: replace Abstract: As LLM-based agents are deployed with increasing autonomy in real-world environments, a critical failure mode emerges: agentic misalignment, where agents choose harmful…
- arXiv cs.AIresearchCan Compact Open-Weight Large Language Models Recover Mobility Events from Incomplete Trajectories? A Controlled Evaluation Using Air Trips
arXiv:2610.24266v1 Announce Type: new Abstract: Incomplete mobility trajectories can obscure the observations needed to identify important travel events and their endpoints. This study evaluates whether compact, locally…
- arXiv cs.AIresearchFrom SEC Filings to Financial Signals: Evaluating LLMs on Stock Market Reactions and Analyst Forecasts
arXiv:2610.24268v1 Announce Type: new Abstract: Form 10-Q MD&A sections combine long-form management narrative with accounting and forward-looking information. Benchmarks for financial filings commonly evaluate question…
- arXiv cs.AIagentsSteerSearch: Interactive Search Reasoning with Mid-Response Steering
arXiv:2610.24287v1 Announce Type: new Abstract: Search-augmented language models increasingly serve as long-horizon research agents, yet most open training methods assume a fixed user goal throughout search. We study mid…
- arXiv cs.AIagentsNot Every User Says It All: Training Tool-Using Agents Beyond Ideal Simulated Users
arXiv:2610.24294v1 Announce Type: new Abstract: Tool-using agents are trained and evaluated almost exclusively with strong simulated users who state every detail clearly, yet real users are often weaker. We construct wea…
- arXiv cs.AIagentsWho Speaks, Who Authorizes, Who Acts? Understanding Trust Boundaries in Multi-Agent LLMs
arXiv:2610.24298v1 Announce Type: new Abstract: Multi-agent LLM systems exchange messages before acting through tools, but message delivery, signing authority, and tool-side enforcement are separate trust boundaries. We…
- arXiv cs.AIresearchCME-SpectrumBench: Can LLMs Analyze Condensed Matter Spectral Data?
arXiv:2610.24304v1 Announce Type: new Abstract: Large Language Models (LLMs) are presently well assessed by expert-level mathematics and physics benchmarks which focus on theoretical, symbolic reasoning. However, we lack…
- arXiv cs.AIresearchThe Judge Can Debias Itself: Awareness and Self-Debiasing for Reliable LLM-as-a-Judge
arXiv:2610.24320v1 Announce Type: new Abstract: LLM-as-a-Judge systems are increasingly employed for automated evaluation, yet remain susceptible to systematic biases that undermine their reliability. We demonstrate that…
- arXiv cs.AIresearchContinuous Ontology Assurance through External Resources -- An Ontology Engineering Method with a Case Study in Software Language Engineering
arXiv:2610.24322v1 Announce Type: new Abstract: Domain ontologies require review against knowledge recorded in evolving external resources. We propose COATER, a method that manages executable review components called arr…
- arXiv cs.AIresearchScreening LLM Judges Without Ground Truth: A Preliminary Evaluation of an AI Dental Hygiene Coach
arXiv:2610.24335v1 Announce Type: new Abstract: Evaluating a fine-tuned coaching model with expert graders is slow and expensive, and an LLM judge cannot be trusted without ground truth. We present a staged pipeline that…
- arXiv cs.AIresearchFlashRT-PAI: Fast Kernels Are Not Fast Pipelines for Physical AI Inference
arXiv:2610.24349v1 Announce Type: new Abstract: Vision-language-action (VLA), video-action (VA), and world action models (WAMs) form the backbone of modern physical AI systems, yet their real-time single-batch inference…
- arXiv cs.AIresearchDKnown-Jev: Fine-Tuning and Evaluation of a JEV-Style Model for Tasks Involving Normative Knowledge
arXiv:2610.24358v1 Announce Type: new Abstract: Tasks involving normative knowledge include concept interpretation, eligibility determination, procedure explanation, material requirements, and service-outcome assessment,…
- arXiv cs.AIresearchA Two-Stage Hybrid Framework for Inferring Air Trips from Mobile Location-based Service (LBS) Data under Incomplete Airport Sightings
arXiv:2610.24366v1 Announce Type: new Abstract: Mobile device location-based service (LBS) data provide large-scale spatiotemporal observations that can help airport planners understand air travelers' mobility patterns.…
- arXiv cs.AIresearchWAQS: Weight Absorption for Quadratic Probing and Affine Steering
arXiv:2610.24367v1 Announce Type: new Abstract: Activation steering provides a lightweight alternative to fine-tuning for controlling large language model behavior. Existing methods, however, often face a trade-off betwe…
- arXiv cs.AIresearchBeyond Next-Screen Prediction: The Decision Value of GUI World Models
arXiv:2610.24372v1 Announce Type: new Abstract: GUI world models are increasingly used to improve action selection by predicting the consequences of candidate actions, but it is unclear how much of the resulting gain com…
- arXiv cs.AIagentsTrain Locally, Learn Together: Distributed Multi-Agent Training via ADMM
arXiv:2610.24380v1 Announce Type: new Abstract: Multi-agent systems (MAS) can outperform individual language-model agents by combining specialized roles, diverse reasoning processes, and collaborative decision-making. Ho…
- arXiv cs.AIagentsIs This Your Agent? Identity Watermarking for Agent Systems in Dynamic Environments
arXiv:2610.24382v1 Announce Type: new Abstract: Agent systems deployed in dynamic environments continuously update their memories, skills, and runtime configurations, leading to evolving trajectories and challenges for p…
- arXiv cs.AIagentsIF-MAP: Influence-Function-Inspired Agent Prioritization in Multi-Agent Systems
arXiv:2610.24383v1 Announce Type: new Abstract: Improving every agent in an LLM-based multi-agent system can be prohibitively costly, making it important to identify which agent should be prioritized to achieve the great…
- arXiv cs.AIagentsBridging Exploration and Persona: Information-Directed Dueling for Test-Time Agent Personalization
arXiv:2610.24385v2 Announce Type: replace Abstract: Effective test-time personalization requires LLM agents to adapt to query-dependent user preferences from lightweight interaction. We study persona selection with binar…
- arXiv cs.AIresearchTrust the Edge, Doubt the Arrow: Auditing Validation Signals for Text-Mined Clinical Causal Knowledge Graphs
arXiv:2610.24390v1 Announce Type: new Abstract: Clinical causal retrieval-augmented generation systems increasingly use knowledge graphs (KGs) mined from notes, yet extracted assertions remain candidates that require aud…
- arXiv cs.AIresearchMSCR: Jointly Balancing Utilization Across Modalities and Discovering Synergistic Information
arXiv:2610.24395v1 Announce Type: new Abstract: In multimodal learning, modality imbalance often lets one modality dominate optimization and keeps the model from fully exploiting cross-modal interactions. Balance-oriente…
- arXiv cs.AIagentsHippo: Training-Free Agent Self-Evolution via Hippocampal Replay
arXiv:2610.24453v2 Announce Type: replace Abstract: Motivated by recursive self-improvement (RSI), we study how LLM agents can learn through external memory without updating model parameters. Extracting useful experience…
- arXiv cs.AIagentsMELD: Recursive Self-Improvement via Hierarchical Coordination for Multi-Agent Scientific Discovery
arXiv:2610.24454v1 Announce Type: new Abstract: Multi-agent scientific discovery with large language models raises a coupled learning problem: research assignments determine both which experiments are pursued and the exp…
- arXiv cs.AIresearchLearning to Compose: Recursive Self-Improvement of Shared Executable Stacks for Zero-Shot Long-Horizon Robot Manipulation
arXiv:2610.24464v1 Announce Type: new Abstract: Individually competent robot skills can fail when composed: tools may lack a required capability, completion signals may misrepresent physical progress, and one skill may l…
- arXiv cs.AIresearchPluralistic Alignment in Large Language Models: Objectives, Human References, and Preference Representations
arXiv:2610.24471v2 Announce Type: replace Abstract: Pluralistic alignment aims to represent meaningful variation in human preferences rather than collapse heterogeneous judgments into a single target. Yet existing work o…
- arXiv cs.AIresearchConCI: A Controlled Counterfactual Benchmark for Contextual Integrity in LLMs
arXiv:2610.24494v1 Announce Type: new Abstract: Contextual integrity (CI) holds that the appropriateness of sharing personal information depends on the context of the information flow, not on the information alone. Howev…
- arXiv cs.AIresearch$M^2Net$: Unifying Scale-Specific Structures and Prototype Decomposition for Spatial-Temporal Forecasting
arXiv:2610.24516v1 Announce Type: new Abstract: Spatial-temporal forecasting is pivotal for intelligent urban management yet remains challenging due to the intricate coupling of spatial dynamics and temporal non-stationa…
- arXiv cs.AIresearchAxiomatic World Modeling for Physics Reasoning
arXiv:2610.24519v1 Announce Type: new Abstract: Recent research-level physics-reasoning benchmarks for LLMs reveal a failure mode that lies not in knowledge or execution, but in modeling the world being reasoned about. M…
- arXiv cs.AIagentsHABIT-Bench: Evaluating Habit Induction from Longitudinal Weak Evidence in Agent Memory
arXiv:2610.24520v2 Announce Type: replace Abstract: In long-term interactions with large language model (LLM) agents, users reveal habits through their choices and actions. These habits capture implicit user information…
- arXiv cs.AIresearchComparing Sparsified MLPs and Transcoders: Many Transcoder Features Are Neurons in Disguise
arXiv:2610.24521v1 Announce Type: new Abstract: MLP neurons in transformer language models are widely understood to be polysemantic, motivating transcoders - sparse, learned approximations of MLP layers - as interpretabl…
- arXiv cs.AIresearchRethinking EEG Spatial Super-Resolution as Dynamic State Decomposition and Filtering
arXiv:2610.24528v1 Announce Type: new Abstract: EEG spatial super-resolution aims to recover high-density neural signals from sparse electrode observations. Existing methods have achieved substantial progress through spa…
- arXiv cs.AIresearchConcordance Plausibility Check for Intervention-Conditioned Human Digital Twins
arXiv:2610.24577v1 Announce Type: new Abstract: Intervention-conditioned human digital twins predict treatment alternatives that cannot be observed in the same patient. Ordinary predictive accuracy therefore cannot estab…
- arXiv cs.AIresearchCADS: Measuring the Incremental Contribution of LLM Proposers in Sequential Biological Screens
arXiv:2610.24597v1 Announce Type: new Abstract: Large language models are increasingly used to guide sequential biological screens, yet standard evaluations rely on raw hit counts that conflate a proposer's value with th…
- arXiv cs.AIagentsHSPR: Recovering Software-Agent Benchmark Scores via Hierarchical Posterior Inference
arXiv:2610.24611v2 Announce Type: replace Abstract: Evaluating software agents on a full benchmark is expensive. A small task subset can preserve agent rankings while yielding inaccurate scores. Historical responses prov…
- arXiv cs.AIagentsSelm: Self-Evolving Library Management for Multi-Skill Agent Optimization
arXiv:2610.24630v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly use skills as a lightweight way to acquire task-specific knowledge. Recent work enables skills to evolve from task trajectori…
- arXiv cs.AIresearchFrequency-Domain Membership Inference Does Not Beat Confidence Thresholding: A Band-Resolved Analysis of Why
arXiv:2610.24643v1 Announce Type: new Abstract: A widely repeated account of membership inference holds that models lean on fine-grained detail for samples they have memorized, so degrading an input's resolution should e…