AI news archive.
4,965 articles filed under Agents, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsCan LLM Agents Automate Reinforcement Learning for Text-to-Speech?
arXiv:2610.04488v1 Announce Type: cross Abstract: Although reinforcement learning (RL) post-training repairs the localized segmental errors of zero-shot text-to-speech (TTS), arriving at a working recipe still relies on…
- arXiv cs.AIagentsQuantifying Collusion Among Autonomous LLM Agents: A Statistical Analysis of the Collusion Wiki Incident
arXiv:2610.04528v1 Announce Type: cross Abstract: In August and September 2026, independent researchers publicly documented an unusual incident: thousands of autonomous agents, self identifying as OpenAI models on web re…
- arXiv cs.AIagentsFrom Probe Scores to Alarm Policies: Operational Validity of Activation Monitors for Language-Model Agents
arXiv:2610.04575v1 Announce Type: cross Abstract: Activation probes can predict safety-relevant properties of language models with high area under the receiver-operating-characteristic curve (AUROC), but deployed agent m…
- arXiv cs.AIagentsRETRACE: From Entangled Repair Histories to Reusable Experience for CI Repair
arXiv:2610.04658v1 Announce Type: cross Abstract: Large language model (LLM) agents increasingly reuse prior experience, but most approaches assume that problems and solutions are already aligned. Software histories rare…
- arXiv cs.AIagentsVoCa: Designing Speech-Canvas Interaction for Voice-Based Conversational Agents
arXiv:2610.04706v1 Announce Type: cross Abstract: People write and sketch while speaking to explain, organize, and develop content together. Inspired by these practices, we investigate how voice agents can use a canvas a…
- arXiv cs.AIagentsOrganising Trajectory Evidence for Language-Model Agent Assurance: Fragments, Methods, and the Residual
arXiv:2610.04710v1 Announce Type: cross Abstract: Methods for assessing language-model agents include rule checkers over logs, analyses of skill coverage and composition, support checkers, prefix monitors, execution gate…
- arXiv cs.AIagentsTowards Automatically Pruning Logging Code with Coding Agents: How Far Are We?
arXiv:2610.04716v1 Announce Type: cross Abstract: Logging code supports debugging, monitoring, and software maintenance, but excessive logging can add noise, impose runtime overhead, and obscure diagnostic information. W…
- arXiv cs.AIagentsARISE: Adaptive Agentic Reasoning with Image-grounded Self-Evaluation for Interpretable IBD Assessment
arXiv:2610.04777v1 Announce Type: cross Abstract: Inflammatory bowel disease (IBD) requires frequent imaging-based assessment, yet interpretation of modalities such as wireless capsule endoscopy (WCE) and intestinal ultr…
- arXiv cs.AIagentsSeparating Decision Time from Decision Quality in the Real-Time Gap of Distilled Deciders: Evidence from a Game and a Conveyor Simulator
arXiv:2610.04810v1 Announce Type: cross Abstract: Real-time agents are often evaluated with the world paused, or with decision time rounded to whole ticks (tick conversion). Tick conversion predicts almost no loss for a…
- arXiv cs.AIagentsViva La Vida: Verification and Accumulation Failures in Multi-Agent Proof Search
arXiv:2610.04829v1 Announce Type: cross Abstract: When an agentic prover works on an open problem, there is no proof assistant to fall back on: its verifier and lemma library are ultimately language models judging model…
- arXiv cs.AIagentsScaling Verifiable Environments for Long-horizon Work Agents
arXiv:2610.04906v5 Announce Type: replace-cross Abstract: Work agents operate over digital artifacts to execute professional knowledge-intensive work, requiring training environments that support long-horizon interaction…
- arXiv cs.AIagentsPreAct-Nav: Agentic Reasoning Before Action for Urban Navigation
arXiv:2610.04916v1 Announce Type: cross Abstract: Urban navigation requires embodied agents to pursue long-horizon goals through local decisions based on egocentric observations. However, existing agentic navigation meth…
- arXiv cs.AIagentsComplex Agents, Shallow Tests: Demystifying and Enhancing Test Adequacy of Agent Harness in the Wild
arXiv:2610.04921v1 Announce Type: cross Abstract: LLM-based agentic systems are emerging as a new software paradigm. Modern agents are typically composed of backbone LLMs and a surrounding harness that serves as the oper…
- arXiv cs.AIagentsAssembling Insights for Agentic Machine Learning Engineering Systems
arXiv:2610.04927v1 Announce Type: cross Abstract: Agentic machine learning engineering (MLE) is an emerging AI4SE application for complex ML tasks and a step toward recursive self-improvement of AI systems. Recent agenti…
- arXiv cs.AIagentsSoftware World Models: From Consequence Prediction to Decision Value
arXiv:2610.04940v1 Announce Type: cross Abstract: A coding agent may safely modify one repository while silently breaking downstream services, libraries, or datastores that depend on it. Exhaustively running integration…
- arXiv cs.AIagentsSparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing
arXiv:2610.05037v1 Announce Type: cross Abstract: Sparse-accelerator design spaces are usually searched against analytical models, so a design point is admitted on what a model predicts rather than on what the hardware d…
- arXiv cs.AIagentsBlocking at the Boundary: Auditing Long-Horizon Agents against Staged Prompt Injection
arXiv:2610.05163v1 Announce Type: cross Abstract: Long-horizon agents consume external content, invoke tools, and modify persistent state. Indirect prompt injection can exploit task-specific context, propagate across cau…
- arXiv cs.AIagentsSettling the Computational Complexity of Max-Min Allocation with Ternary Valuations
arXiv:2610.05237v1 Announce Type: cross Abstract: We study the problem of computing an allocation of indivisible items that maximizes egalitarian welfare, i.e., the utility of the worst-off agent, when agents' item value…
- arXiv cs.AIagentsStateWise: Diagnosing and Repairing Persistent Operational State Before Agent Actions
arXiv:2610.05241v1 Announce Type: cross Abstract: LLM agents combine reasoning, tool use, and persistent memory to support work across tasks by reusing stored operational records as premises for later actions. However, e…
- arXiv cs.AIagentsWho Is Your Agent Serving? Provider-Side Indirect Prompt Injection in Proactive Agents
arXiv:2610.05266v1 Announce Type: cross Abstract: Proactive personal agents increasingly decide what to recommend, how to personalize advice, and what follow-up assistance to offer, creating a new user-decision attack su…
- arXiv cs.AIagentsIRSTD-Agent: Agentic Infrared Small Target Detection via Zoom-Guided Interaction Learning
arXiv:2610.05342v1 Announce Type: cross Abstract: Infrared small-target detection plays an important role in maritime monitoring and aerial surveillance. Although multimodal large language models (MLLMs) offer promising…
- arXiv cs.AIagentsAIProver: Agentic Auto-Formalization of Mathematical Research via Certificate-Driven Evolving Harness
arXiv:2610.05367v1 Announce Type: cross Abstract: Proof auto-formalization translates natural-language (NL) theorems and proofs into a formal language (FL) such as Lean, enabling mechanical verification. Despite rapid pr…
- arXiv cs.AIagentsG-CARB: Graph-Localized Conformal Agent Risk Budget for Compositional Harm
arXiv:2610.05563v1 Announce Type: cross Abstract: Small language model (SLM) agents need safety controls that track consequences across tool calls with little monitoring overhead. A private read, for example, becomes a l…
- arXiv cs.AIagentsScenario-Based Compositional Statistical Model Checking for Safety Specifications
arXiv:2610.05571v1 Announce Type: cross Abstract: In safety-critical domains such as autonomous driving, systems must be evaluated across a large number of environment conditions, often represented as composite scenarios…
- arXiv cs.AIagentsAgentDoxx: Agentic Re-identification of Anonymized Text with Web Search
arXiv:2610.05586v1 Announce Type: cross Abstract: As Large Language Models (LLMs) gain tool use capabilities such as web search, they can retrieve and cross-reference public information, creating privacy risks beyond mem…
- arXiv cs.AIagentsUndoBench: Separating Task Competence from Recovery Capability in Tool-Using AI Agents
arXiv:2610.05622v1 Announce Type: cross Abstract: Tool-using AI agents are increasingly deployed across enterprise software systems, yet widely used benchmarks primarily evaluate nominal task completion, conflating basel…
- arXiv cs.AIagentsNexus: An Execution Fabric for AI Agents Across Cloud, Edge, and Devices
arXiv:2610.05709v1 Announce Type: cross Abstract: Language-model agents are evolving into long-running services that interact with models, tools, computers, mobile devices, and distributed environments. Existing agent fr…
- arXiv cs.AIagentsBeyond Semantic Similarity: Performance and Costs of Agentic Retrieval for Complex Tasks
arXiv:2610.05750v1 Announce Type: cross Abstract: Modern information systems, including many agentic workflows, use dense retrieval to explore large amounts of unstructured data. However, dense retrieval relies on surfac…
- arXiv cs.AIagentsWho Keeps the Gains from Personal AI Assistants? Seller Adaptation and the Unassisted in a Language-Model Market Simulation
arXiv:2610.05823v1 Announce Type: cross Abstract: Personal AI assistants are beginning to transact for consumers, and early adopters capture real savings. Whether those savings survive, and what happens to consumers who…
- arXiv cs.AIagentsSelecting Long-Horizon Trajectories for Reliable and Efficient Terminal-Agent Training
arXiv:2610.05831v1 Announce Type: cross Abstract: Terminal agents are commonly trained by imitating long teacher trajectories, yet how much of each trajectory to supervise remains unexplored. We study the \emph{supervisi…
- arXiv cs.AIagentsImagine to Act: High-Fidelity Data Synthesis via Image Editing World Model for Scalable GUI Agent Training
arXiv:2610.05861v1 Announce Type: cross Abstract: Graphical User Interface (GUI) agents have emerged as a promising paradigm for automating complex digital workflows across diverse applications. However, training highly…
- arXiv cs.AIagentsThunderSyncRL: Lossless Acceleration of Agentic Reinforcement Learning
arXiv:2610.05935v1 Announce Type: cross Abstract: Language models are moving beyond generating answers to pursuing long-horizon goals in interactive environments. Post-training these agents requires long, heterogeneous t…
- arXiv cs.AIagentsRunaway Reaction: When Benign Skills Compose into Malicious Behavior
arXiv:2610.05943v1 Announce Type: cross Abstract: Agent skills package task-specific knowledge and procedures that can be composed to support complex agent tasks, while public marketplaces provide a growing pool of reusa…
- arXiv cs.AIagentsAgentSpy: Making AI Agent Behavior Observable
arXiv:2610.06001v1 Announce Type: cross Abstract: AI agents built on large language models (LLMs) run shell commands, read and write files, and reach the network, typically with their user's privileges. However, what an…
- arXiv cs.AIagentsParallelism or Concession? Concurrency-Aware Procurement Negotiation for Agentic Commerce
arXiv:2610.06017v1 Announce Type: cross Abstract: Agentic buyers can cheaply fork a procurement task into many parallel negotiations, but concurrency is not free: every thread consumes resources, and simultaneous agreeme…
- arXiv cs.AIagentsAttention Tax, Handoff Tax: A Stylised Model of When Multi-Agent LLM Systems Help
arXiv:2610.06069v1 Announce Type: cross Abstract: Recent work on multi-agent LLM systems reaches sharply different conclusions: some results show that a single agent with the same information and compute should dominate…
- arXiv cs.AIagentsWhere Did the Repair First Go Wrong? Localizing the Origins of Silent Failures in Agentic Vulnerability Repair
arXiv:2610.06163v1 Announce Type: cross Abstract: Localizing where an LLM-based agent first fails to uphold security during a repair can show which stage of its workflow needs an additional safeguard. This is difficult f…
- arXiv cs.AIagentsCorrect Code, Broken Contributions? SWE-CC: Benchmarking Repository Policy Compliance for Coding Agents
arXiv:2610.06193v1 Announce Type: cross Abstract: Autonomous coding agents now resolve a substantial share of real-world GitHub issues. However, passing functional tests differs fundamentally from producing a high-qualit…
- arXiv cs.AIagentsAPOD: reasoning-guided agentic population ordinary differential equation discovery for pharmacological digital twins
arXiv:2610.06227v1 Announce Type: cross Abstract: Establishing ordinary differential equations (ODEs) describing population data is a fundamental part of mathematical modeling in pharmacology, crucial to developing digit…
- arXiv cs.AIagentsAgentic schema-guided extraction of materials process knowledge from scientific literature
arXiv:2610.06322v1 Announce Type: cross Abstract: Materials literature contains detailed experimental knowledge, but procedures, chemical entities and measurements remain difficult to aggregate because they are reported…
- arXiv cs.AIagentsRAISED: Self-Distillation for Robustness to Prompt Injection in LLM Agents
arXiv:2610.06401v2 Announce Type: replace-cross Abstract: Tool-using language-model agents are vulnerable to indirect prompt injection because they must act on untrusted external content. Existing training-time defenses…
- arXiv cs.AIagentsValid Stopping in Adaptive Generator-Verifier Loops
arXiv:2610.06432v1 Announce Type: cross Abstract: Numerous agentic workflows are based on a generator-verifier loop: a generator proposes candidates, a cheap verifier scores them, and the workflow terminates when a propo…
- arXiv cs.AIagentsGCTAuto-encoder: A Cross modal Framework for Security Flaw Detection in IoT Networks
arXiv:2610.06517v1 Announce Type: cross Abstract: IoT encompasses diverse physical entities, from smart home devices to autonomous vehicles, creating a complex environment with heterogeneous security models. This heterog…
- arXiv cs.AIagentsDoes AI Help Cyber Attackers or Defenders? Evidence from Nonpublic Vulnerabilities and Subsequent Attacks
arXiv:2610.06584v1 Announce Type: cross Abstract: The release decision for frontier AI systems increasingly relies on cyber capability benchmarks, yet public vulnerability benchmarks can expose agents to previously publi…
- arXiv cs.AIagentsMind the Accent Gap: British Accent Robustness in Speech-Driven Financial Voice Assistants
arXiv:2610.06587v1 Announce Type: cross Abstract: AI voice assistants often use Automatic Speech Recognition (ASR) with LLM-based reasoning, yet existing systems struggle with regional British accents, including Scottish…
- arXiv cs.AIagentsWord-Level Text Unmixing via Evidence-Preserving Ownership Routing with Language Models
arXiv:2610.06603v1 Announce Type: cross Abstract: Text from multiple sources can become interleaved into a single sequence when attribution metadata is lost, such as overlapping speech transcripts, document reading flows…
- arXiv cs.AIagentsLarge Language Model-Guided Discovery of Weight-Five Bivariate Bicycle Codes
arXiv:2610.06623v1 Announce Type: cross Abstract: Building on our earlier program-evolution workflow guided by large language models (LLMs), we study weight-five bivariate bicycle (BB) and perturbed bivariate bicycle (PB…
- arXiv cs.AIagentsVideoTapestry: Query-Adaptive Memory Refinement for Multi-Agent Long-Video Understanding
arXiv:2610.06672v1 Announce Type: cross Abstract: Long-video understanding places substantial demands on memory, as answering questions often requires retrieving information distributed across extended temporal spans. Ex…
- arXiv cs.AIagentsBazaarBench: Delegation Safety in Decentralized C2C Marketplaces Run by LLM Agents
arXiv:2610.06748v1 Announce Type: cross Abstract: In decentralized consumer-to-consumer (C2C) marketplaces, people list goods, negotiate with strangers, and rate one another, so trust rests on reputation. Large language…
- arXiv cs.AIagentsCLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling
arXiv:2610.06829v1 Announce Type: cross Abstract: Open-source web agents are now strong enough to execute realistic browser tasks, but training them with reinforcement learning still depends on weak supervision: binary t…