AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsContextRender: From Execution Dependencies to Agent Context
arXiv:2609.37743v1 Announce Type: new Abstract: LLM agents performing long-horizon tasks accumulate tool results that later steps may need. Passing the full history to every invocation is costly even when it fits within…
- arXiv cs.AIresearchCross-Entropy Guided Routing in Mixture-of-Experts Large Language Models
arXiv:2609.37751v1 Announce Type: new Abstract: Sparse mixture-of-experts (MoE) large language models scale model capacity by routing each token to a small subset of experts. Their routers are regularized with load balan…
- arXiv cs.AIagentsOmniVCBench: Benchmarking Evidence-Grounded Multimodal Reasoning Towards AI Virtual Cells
arXiv:2609.37773v1 Announce Type: new Abstract: Artificial Intelligence Virtual Cells (AIVCs) are envisioned as scientific agents that simulate cellular responses, explain underlying mechanisms, and support hypothesis-dr…
- arXiv cs.AIresearchAdam under Generalized Smoothness with Second-Moment-Type Stochastic Gradients
arXiv:2609.37787v2 Announce Type: replace Abstract: Adam is widely observed to remain stable even when the objective deviates significantly from global smoothness. Under the generalized smoothness framework, however, exi…
- arXiv cs.AIresearchA neural network that maintains and retrieves memories based on context
arXiv:2609.37791v1 Announce Type: new Abstract: Every day, people continuously infer situational context and adjust the way they understand and remember the world. Context, signaled by the prefrontal cortex, is known to…
- arXiv cs.AIresearchDIET: Deletion-response Expert Trimming for Video Diffusion Transformers
arXiv:2609.37829v1 Announce Type: new Abstract: Video diffusion transformers (DiTs) increasingly adopt mixture-of-experts (MoE) architectures to reduce active computation, but their full expert storage remains costly. Ex…
- arXiv cs.AIresearchCan a Cacheable Decision Model Follow Rules?
arXiv:2609.37832v1 Announce Type: new Abstract: Certo is a small non-generative decision model (Qwen3-4B): it scores candidate actions from their text and returns a probability, instead of generating an answer. The accur…
- arXiv cs.AIagentsMixture of Self-Improving Branches For Agent Harness Optimization
arXiv:2609.37834v1 Announce Type: new Abstract: Harness optimization provides a practical setting for recursive self-improvement (RSI), where agent-generated modifications inform subsequent changes through execution feed…
- arXiv cs.AIresearchActive Budget Can Kill Sensitivity: Diagnosing and Repairing TopK Sparse Autoencoder Reliability
arXiv:2609.37857v2 Announce Type: replace Abstract: Sparse autoencoders (SAEs) are increasingly scaled to wider dictionaries to recover fine-grained structure from large language model activations. However, a feature is…
- arXiv cs.AIresearchCo-PiLOT: Constrained Physics-Informed Latent Optimization for Target-Driven Inverse Design
arXiv:2609.37875v1 Announce Type: new Abstract: Inverse design of physical systems (molecules, devices, microstructures) often reduces to optimizing a high-dimensional structure against an expensive black-box simulator.…
- arXiv cs.AIagentsGuide, Then Let Go: Gap-Adaptive Teacher Scheduling for Sparse-Reward Agentic RL
arXiv:2609.37898v1 Announce Type: new Abstract: Reinforcement learning for long-horizon agents typically relies on sparse outcome-based rewards. This leads to a severe cold-start problem, as early-stage policies often fa…
- arXiv cs.AIresearchYou Cannot Pick a Provider From the Price List: Market-Aware Routing for Open-Weight LLM Inference
arXiv:2609.37902v1 Announce Type: new Abstract: Existing LLM routers choose among models using static per-model costs. We show that open-weight inference markets introduce a second, largely ignored decision axis: after c…
- arXiv cs.AIagentsPixels to Keys: Exploring Spatial and Motion Cues in Gameplay Inverse Dynamics
arXiv:2609.37907v2 Announce Type: replace Abstract: Video games offer scalable environments for studying perception and control in embodied agents. Abundant online gameplay videos could supply demonstrations, but they ra…
- arXiv cs.AIresearchGRFBrain: Graph-Structured Rectified Flows for EEG Dynamic Modeling
arXiv:2609.37934v1 Announce Type: new Abstract: Forecasting time-varying functional connectivity from electroencephalography (EEG) requires modeling both history-dependent trends and structured variability across channel…
- arXiv cs.AIagentsVideo-RSI: Recursive Self-Improvement of Video Understanding Agents via Harness Evolution
arXiv:2609.37950v1 Announce Type: new Abstract: Video understanding agents acquire evidence through an executable harness that controls what they observe and how they use those observations. However, execution traces con…
- arXiv cs.AIagentsTopological Coherence for Self-evolving Multi-agent Systems
arXiv:2609.37953v1 Announce Type: new Abstract: Complex tasks inherently couple workflow structure, agent responsibility, collaboration, and memory access: task regions delimit responsibility and tool scope, cross-region…
- arXiv cs.AIresearchBrainNet Studio: A Unified Toolkit for Brain Network Construction, Intelligent Analysis, and Visualization
arXiv:2609.37956v1 Announce Type: new Abstract: Brain networks characterize structural and functional relationships among brain regions and support research on cognition, brain disorders, and brain-computer interfaces. T…
- arXiv cs.AIagentsSelfSearch: Reward-Free Search for Self-Improving Agents
arXiv:2609.37968v2 Announce Type: replace Abstract: Advances in the coding capabilities of LLM agents allow them to inspect and modify their own instructions, tools, and execution procedures. Existing approaches use this…
- arXiv cs.AIresearchKV-Kaizen: Learning Context-Adaptive Cache Compression Choices
arXiv:2609.37988v2 Announce Type: replace Abstract: As the context size of text processed with an LLM grows, the size of KV caches can outstrip the memory allocated for the original model weights. This impacts LLM throug…
- arXiv cs.AIresearchWhich Attention Heads are like the Human Head? Not the Ones that Compute
arXiv:2609.37991v1 Announce Type: new Abstract: Brain-AI alignment is often interpreted as a sign that model and brain perform similar computations. Whether the aligned units are causally involved in model computation is…
- arXiv cs.AIagentsDiagnosing and Improving Probabilistic Reasoning in Large Language Models
arXiv:2609.38005v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly proposed as decision assistants who must reason probabilistically from available evidence under explicit decision costs. W…
- arXiv cs.AIresearchHARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model Deployment
arXiv:2609.38006v1 Announce Type: new Abstract: Running a language model locally offers advantages in privacy, latency, and cost, but local hardware fits only small models, which are less capable than frontier models. Th…
- arXiv cs.AIagentsBrain-SAD: A Brain-Inspired Safe Autonomous Driving Control Framework with Dynamic Fear-Oriented Constraint on Dual-Policy
arXiv:2609.38016v1 Announce Type: new Abstract: Constrained Reinforcement Learning has recently gained increasing attention in the field of Safe Autonomous Driving, where the general mechanism is to maximize the expected…
- arXiv cs.AIresearchPE-EK-PINN: Physics Embedding with Evolving Kernel for Scalable Physics-Informed Neural Networks
arXiv:2609.38023v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) embed governing equations into deep learning, but enforce them only through loss residuals, leaving highly oscillatory wave behavio…
- arXiv cs.AIagentsRetrieval-Augmented Skill Optimization via Cross-Harness Adaptation
arXiv:2609.38024v1 Announce Type: new Abstract: An agent skill is a reusable, actionable natural-language artifact that guides an agent to perform a task effectively under a given harness. Recent studies have explored th…
- arXiv cs.AIagentsUserProxyBench: Evaluating LLM User Simulators for Agent Benchmarks and Training
arXiv:2609.38043v1 Announce Type: new Abstract: Interactive agent benchmarks and multi-turn reinforcement learning increasingly place a second language model in the role of the user. This simulated user controls what inf…
- arXiv cs.AIresearchProbability is Not Enough: Exploring and Counting Divergent Tokens for Reasoning Uncertainty Quantification in LLMs
arXiv:2609.38070v1 Announce Type: new Abstract: As the chain-of-thought reasoning capabilities of large language models improve, evaluating and calibrating their reasoning confidence is becoming increasingly important fo…
- arXiv cs.AIagentsCharacter Training for Risk-Averse Agents
arXiv:2609.38093v1 Announce Type: new Abstract: Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm. Misaligned but risk-averse agents would tend to favor safer strategies like ma…
- arXiv cs.AIresearchNeuronEye: Query-Guided Visual Concept Activation for Vision-Language Reasoning
arXiv:2609.38098v1 Announce Type: new Abstract: Current vision-language models (VLMs) encode visual information in dense hidden states where object identity, spatial layout, and local attributes are implicitly entangled…
- arXiv cs.AIagentsDo LLM Agents Execute the Plans They Declare? From Planning-Mode Declaration to Pattern-Specific Execution
arXiv:2609.38108v1 Announce Type: new Abstract: Large language models (LLMs) enable agents to solve long-horizon tasks by generating a plan and then executing it in an environment. However, successful planning requires t…
- arXiv cs.AIresearchStochastic World Models for Verifying Vision-Based Neural Feedback Systems
arXiv:2609.38120v1 Announce Type: new Abstract: Verifying a vision-based neural feedback system requires a model of the observations its controller acts upon. Such a model must capture the variation the sensor produces,…
- arXiv cs.AIresearchAdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation
arXiv:2609.38142v1 Announce Type: new Abstract: A small trainable advisor can steer a frozen language-model executor using natural-language advice. In addition to learning from task rewards, the advisor can use feedback…
- arXiv cs.AIagentsLearning Meta-Skills for Agent Harness Design in Test-Time AI4AI
arXiv:2609.38143v2 Announce Type: replace Abstract: Agent performance depends on both reasoning ability and the environment in which it acts. We study test-time AI-for-AI, asking how a Builder can learn to construct bett…
- arXiv cs.AIagentsThinking Before Thinking: Scaling Agentic Inference Through Meta-Reasoning
arXiv:2609.38147v1 Announce Type: new Abstract: As agents take on longer and more complex problems, controlling the execution becomes a task in its own right. Each step in the run brings new control choices, like which p…
- arXiv cs.AIresearchBathtubs, Boundaries, and Sandboxes: AI Regulatory Learning under Legal Uncertainty
arXiv:2601.04094v4 Announce Type: cross Abstract: Effective regulation of AI is a defining policy challenge, driven by their integration into all aspects of society. To remain responsive to their rapid development and em…
- arXiv cs.AIresearchSage: Formalization with Semantic Correction
arXiv:2609.35790v2 Announce Type: replace-cross Abstract: While neural theorem provers have achieved impressive milestones in formal mathematics, they largely operate on the assumption that faithful Lean 4 formal stateme…
- arXiv cs.AIresearchCalibration-First Cross-Cohort Multimodal Temporal Learning for Transferable Asthma-Risk Forecasting
arXiv:2609.35795v1 Announce Type: cross Abstract: Asthma deterioration forecasting must remain reli- able when patient populations, sensor ecosystems, and available modalities change across cohorts. Existing models commo…
- arXiv cs.AIresearchBinarization Flattens the Score Space
arXiv:2609.35797v2 Announce Type: replace-cross Abstract: Large language model (LLM) judges are often used as rewards to train policies on objectives that deterministic verifiers cannot capture. However, these rewards ar…
- arXiv cs.AIagentsEvaluating the Effects of Prompt Perturbation on Bias and Hallucination in Large Language Models
arXiv:2609.35804v1 Announce Type: cross Abstract: Large language models (LLMs) have shown remarkable capabilities in various natural language processing tasks, leading to their widespread deployment as intelligent assist…
- arXiv cs.AIresearchAlignment Forecasting: Predicting Misalignment From Training Data
arXiv:2609.35805v2 Announce Type: replace-cross Abstract: Training a language model on data with a narrow flaw can sometimes make the model broadly misaligned. Inspecting the data at face value often does not settle whet…
- arXiv cs.AIagentsFrom Lexical Baselines to Agentic Retrieval-Augmented Generation: Structured Skill and Responsibility-Level Extraction with the SFIA Framework
arXiv:2609.35806v1 Announce Type: cross Abstract: Automated skill extraction underpins workforce planning, yet most systems represent skills as flat labels with no notion of the responsibility level at which a skill is p…
- arXiv cs.AIagentsEnvironment Steering: Using Data Flow Control to Improve Agent Utility and Safety
arXiv:2609.35807v1 Announce Type: cross Abstract: LLM agents can make unsafe tool calls even when instructed to behave safely. Existing defenses constrain agents before execution, modify tool inputs/outputs, or rely on L…
- arXiv cs.AIagentsWhen Successful Memories Mislead Embodied Agents:Memory Adaption For Task-Conditioned Execution
arXiv:2609.35808v1 Announce Type: cross Abstract: Experience reuse can reduce repeated exploration in embodied agents, but a trajectory that succeeded previously may be unsuitable for the current execution context. Exist…
- arXiv cs.AIresearchCan Multimodal Large Language Models Generate and Detect Multimodal Social Media Fake News?
arXiv:2609.35809v1 Announce Type: cross Abstract: The rapid advancement of generative AI raises concerns about the misuse of Multimodal LLMs (MLLMs) for large-scale disinformation campaigns on social media. Despite exist…
- arXiv cs.AIresearchTRACE: Deployable Tree-Relational Structure Enhancement for Oncology LLMs
arXiv:2609.35810v1 Announce Type: cross Abstract: Large language models are increasingly used in oncology applications, but their predictions are often weakly grounded in explicit medical structure. We present TRACE, a d…
- arXiv cs.AIagentsLookahead-R: Budget-Aware Tool Retrieval via Execution-Centric Planning
arXiv:2609.35811v1 Announce Type: cross Abstract: Tool retrieval is a critical bottleneck for LLM-based agents operating over large, heterogeneous API ecosystems. Existing approaches face an inherent trade-off: semantic…
- arXiv cs.AIagentsLocal Predictability and Collective Fidelity in LLM-Agent Societies
arXiv:2609.35813v1 Announce Type: cross Abstract: Compact surrogates could reduce the cost of simulating large language model societies, but must reproduce collective behavior. We compare individual predictions and colle…
- arXiv cs.AIagentsConstructing Challenging Browser-Use Tasks by Controlled Environment Interventions
arXiv:2609.35814v2 Announce Type: replace-cross Abstract: As browser-use agents improve, benchmarks keep pace by adding new tasks, websites, and applications. Doing so is costly, and because many factors change at once,…
- arXiv cs.AIresearchHow to Run Statistics over LLM Judges and Trust the Results: Calibrated Inference for Small-Sample AI Evaluation with evalstats
arXiv:2609.35815v1 Announce Type: cross Abstract: Researchers across academia increasingly base significance claims on LLM judge scores and small-sample AI evaluations. Yet without well-calibrated confidence intervals (C…
- arXiv cs.AIagentsPrimeSeeker: Capability-Oriented Supervision for Deep Search Agents
arXiv:2609.35816v1 Announce Type: cross Abstract: Large language model search agents are often trained with synthetic questions whose difficulty is increased through larger evidence graphs, additional hops, and longer tr…