AI news archive.
4,966 articles filed under Agents, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsCREDO: Variance-Guided Rubric Evolution for Replay-Corrected Credit Assignment
arXiv:2609.24174v1 Announce Type: new Abstract: Long-horizon language agents receive sparse terminal feedback, while intermediate rubrics provide structured but potentially misspecified assessments of progress. In resett…
- arXiv cs.AIagentsHow Many Pixels Is a Digit Worth? Place-Aware Coordinate Entropy for GUI Agent Confidence Estimation
arXiv:2609.24277v1 Announce Type: new Abstract: GUI agents predict click coordinates as digit-token sequences, but standard text-LLM confidence estimation methods rank correct clicks from wrong ones only weakly. GUI-spec…
- arXiv cs.AIagentsWhen and How Should an Agent Clarify? CIGAsk: Teaching LLMs to Clarify via Counterfactual Information Gain
arXiv:2609.24290v1 Announce Type: new Abstract: Instruction-tuned LLMs faced with underspecified queries often commit to a single interpretation rather than ask for clarification, producing confidently wrong answers. In…
- arXiv cs.AIagentsFew-Shot Demonstrations Elicit the Use of In-Context World Representations in LLMs
arXiv:2609.24352v1 Announce Type: new Abstract: Large language models (LLMs), when acting as agents, are expected to take observed data in context, infer the latent state space underlying the world, and leverage it for d…
- arXiv cs.AIagentsVLM-in-Sandbox: Visual Workspaces for Agentic Visual Reasoning
arXiv:2609.24362v1 Announce Type: new Abstract: Sandboxed computer environments support multi-step reasoning with tools, executable programs, and persistent files, yet their extension from language models to vision-langu…
- arXiv cs.AIagentsAscent: An Agentic System over the Model Context Protocol for Real-World Clinical Data Analysis
arXiv:2609.24620v1 Announce Type: new Abstract: Answering epidemiological questions from real-world clinical data requires medical coding, schema-aware SQL, and validation of implicit choices about populations, denominat…
- arXiv cs.AIagentsDUMA-Bench: A Dual-Control Multi-Agent Benchmark for Evaluating LLM Agent Security
arXiv:2609.24662v1 Announce Type: new Abstract: LLM-based agents increasingly operate in environments where they interact with users, tools, and external systems. Yet most security evaluations assume passive users and st…
- arXiv cs.AIagentsBeyond Endpoint Performance: Process-Level Evaluation of Self-Evolving Agents
arXiv:2609.24663v1 Announce Type: new Abstract: Self-evolving agents convert interaction feedback into persistent artifacts, such as memories or skills, which in turn guide subsequent decisions. As these artifacts are it…
- arXiv cs.AIagentsWorld State Generator
arXiv:2609.24744v1 Announce Type: new Abstract: Language agents solve complex tasks through plans and actions. A single step the world refuses puts the goal out of reach, and what the agent does next decides the task. Pr…
- arXiv cs.AIagentsEpi-Logic: A Conceptual Framework for Epistemic Runtime Control, Schema Validity Checking, and Controlled Accommodation in Autonomous AI Agents
arXiv:2609.24755v1 Announce Type: new Abstract: Autonomous AI agents are increasingly deployed in areas where wrong decisions are hard to reverse. This paper examines schema mismatch: the condition in which an agent oper…
- arXiv cs.AIagentsGRUET: Quantifying Uncertainty of Agentic Reasoning-and-Acting Processes
arXiv:2609.24831v1 Announce Type: new Abstract: Agents have attracted considerably increasing attention due to the power of executing both Reasoning and Acting (ReAct) in open and dynamic environments. The ReAct process…
- arXiv cs.AIagentsMedRSI: Recursive Self-Improvement for Medical Agents via Clinically Aligned Self-Evolution
arXiv:2609.24838v1 Announce Type: new Abstract: Medical agents increasingly combine general reasoning models with specialized clinical tools, yet their capabilities remain largely fixed by what clinicians and engineers d…
- arXiv cs.AIagentsPartner-Specific Affective Precision in Social Active Inference
arXiv:2609.24876v1 Announce Type: new Abstract: In multi-agent social settings, model reliability varies across relationships. Beyond inferring what others will do, an agent must calibrate how confidently those inference…
- arXiv cs.AIagentsEt Tu, Brute? Economic Misalignment in Personal AI Agents
arXiv:2609.24927v2 Announce Type: replace Abstract: Personal AI agents make recommendations and take actions on people's behalf in high-stakes economic contexts, e.g., buying a flight, choosing health insurance, or selec…
- arXiv cs.AIagentsEmergent Collusion in Long-Horizon LLM Agent Interaction
arXiv:2609.24967v2 Announce Type: replace Abstract: LLM agents are increasingly deployed in collaborative settings, yet long-term interaction may give rise to undesirable coordination. We study the emergence of collusion…
- arXiv cs.AIagentsHarness-Zero: Harness Distillation via Agent-as-Harness
arXiv:2609.24974v1 Announce Type: new Abstract: Agent harnesses, the external systems that mediate model-environment interaction, can substantially improve agent performance, but their gains remain tied to the harness at…
- arXiv cs.AIagentsBeyond Task Completion: Training Capable and Safe Computer-Use Agents
arXiv:2609.22178v2 Announce Type: replace-cross Abstract: Computer-use agents (CUAs) have made rapid progress in completing complex tasks through graphical user interfaces, yet post-training centered on task success alon…
- arXiv cs.AIagentsA Pinch of SFT, A Dash of RL: When Reinforcement Learning Helps Long-Horizon Advertising Agents
arXiv:2609.22194v1 Announce Type: cross Abstract: Enterprise analytics agents solve long-horizon tool-use problems over distributed business data, requiring retrieval, reasoning, API calls, code execution, and adaptation…
- arXiv cs.AIagentsEvoRank: LLM-Guided Evolution of Multi-Objective Learning-to-Rank Pipelines
arXiv:2609.22196v1 Announce Type: cross Abstract: We present EvoRank, an open autonomous ranking engineer: an LLM-guided evolutionary loop that discovers complete Learning-to-Rank pipelines (features, models, losses, ens…
- arXiv cs.AIagentsPII-TRACE: A Benchmark for Context-Aware PII Detection in Multi-Turn LLM Conversations
arXiv:2609.22200v1 Announce Type: cross Abstract: LLM assistants and agentic systems log long multi-turn conversations. AI providers often scan these conversations for Personally Identifiable Information (PII) and mask t…
- arXiv cs.AIagentsToollery: Scaling LLM Agents to Thousands of Skills and Tools
arXiv:2609.22218v2 Announce Type: replace-cross Abstract: As LLM agents are exposed to hundreds to tens of thousands of skills, tools, and API functions, full-library prompting becomes costly, slow, and less reliable: ea…
- arXiv cs.AIagentsCan Coding Agents Reproduce Official Statistics? Metadata, Retry Budget and the Limits of Execution Feedback in a Controlled Eurostat Benchmark
arXiv:2609.22222v1 Announce Type: cross Abstract: Large language models can generate executable data-analysis code, but successful execution is not equivalent to a valid official-statistics result. This study asks whethe…
- arXiv cs.AIagentsH2LooP Telecom Model v1: From Telecom Comprehension to Autonomous Issue and PR Resolution
arXiv:2609.22241v1 Announce Type: cross Abstract: We present H2LooP Telecom Model v1, a domain-specialized large language models fine-tuned for the telecommunications industry. We release two domain-adapted model variant…
- arXiv cs.AIagentsUniversal Observatory Graphs for Distributed Sky Coverage and Artificial Intelligence Based Interplanetary Routing
arXiv:2609.22244v1 Announce Type: cross Abstract: This research proposes the Universal Observatory Graph (UOG), an AI-driven framework for distributed astronomical observation across the Solar System. The proposed archit…
- arXiv cs.AIagentsCHART: A Harness-Rotation Curriculum for Harness-Robust Search Agents
arXiv:2609.22247v1 Announce Type: cross Abstract: Search agents are usually trained under a single harness. But once an agent is deployed in a real application, its harness is frequently updated (e.g., a rewritten system…
- arXiv cs.AIagentsA Tutorial on Prompt Engineering: From Messy Thoughts to AI Workflows
arXiv:2609.22249v1 Announce Type: cross Abstract: This paper treats prompt engineering as a discipline for turning informal human intent into structured AI work specifications. It develops the practice as a sequence of r…
- arXiv cs.AIagentsPredictors and Orchestrators: Parsimonious Machine Learning within an Agentic AI Harness for Multi-Horizon Karst Aquifer Forecasting
arXiv:2609.22251v1 Announce Type: cross Abstract: Forecasting karst aquifer dynamics is difficult because recharge responses are nonlinear, event-driven, and governed by strongly heterogeneous flow paths. This study deve…
- arXiv cs.AIagentsDeep Persona: A Psychologically Grounded Architecture and Evaluation Framework for Role-Playing Agents and Simulations
arXiv:2609.22255v1 Announce Type: cross Abstract: Existing approaches to persona simulation with Large Language Models (LLMs) mostly rely on shallow character descriptions that fail to sustain coherent character behavior…
- arXiv cs.AIagentsRS-Claw-Evolution: Environment-Feedback-Driven Evolution for Lightweight Remote Sensing Agents in Long-Horizon Tasks
arXiv:2609.22258v1 Announce Type: cross Abstract: Large language model-driven remote sensing (RS) agents offer a promising approach to automating geospatial analysis. However, lightweight RS agents based on compact langu…
- arXiv cs.AIagentsWhich Part of the Context Layer Does the Work? Separating Semantic Content from Retrieval Scaffolding in Text-to-SQL Agents
arXiv:2609.22259v1 Announce Type: cross Abstract: Context layers, curated documentation that an analytics agent fetches at query time, produce large accuracy gains on text-to-SQL benchmarks. A with/without comparison can…
- arXiv cs.AIagentsAuthority-Preserving Evaluation of Medical Vision-Language Assistants
arXiv:2609.22302v1 Announce Type: cross Abstract: Medical vision-language models can propose how urgently a skin lesion should be reviewed, but the local service retains authority to accept or replace that proposal under…
- arXiv cs.AIagentsGameReplica: A Benchmark for Black-Box Visual Game Replication by Vision-Language Agents
arXiv:2609.22308v1 Announce Type: cross Abstract: Coding-agent benchmarks usually evaluate implementation after the target behavior has been specified in text, code, or demonstrations. Existing research has extensively e…
- arXiv cs.AIagentsInitial Evaluation of Potential Bias in Coverage of Humans in Wikidata
arXiv:2609.22375v2 Announce Type: replace-cross Abstract: Introduction. Open collaborative knowledge graphs such as Wikidata increasingly ground agentic artificial intelligence, information retrieval, and language modeli…
- arXiv cs.AIagentsA Governance-Aware Large Language Model Orchestrated Agentic Digital Twin for Transmission System Operator Control Room Decision Support
arXiv:2609.22476v1 Announce Type: cross Abstract: Transmission system operators face rising complexity from renewable integration, reduced inertia, and tighter security margins. Large language models offer natural-langua…
- arXiv cs.AIagentsZero-Trust Authorization and Discovery for Enterprise MCP
arXiv:2609.22573v1 Announce Type: cross Abstract: LLM agents translate natural-language context, which may include attacker-controlled text, into privileged tool calls, so authorization must remain effective even when an…
- arXiv cs.AIagentsFairly Compensated Distributed Information Retrieval and Augmentation for AI Agents
arXiv:2609.22601v1 Announce Type: cross Abstract: The increasing reliance of autonomous AI agents on external and distributed knowledge sources introduces a fundamental challenge for decentralized information marketplace…
- arXiv cs.AIagentsFrom Capability to Assurance in Autonomous Penetration-Testing Harnesses: A Framework and Reference Implementation
arXiv:2609.22664v1 Announce Type: cross Abstract: Research on large language model agents for penetration testing is evaluated almost entirely by capability: whether the agent captures a flag or reproduces a proof of con…
- arXiv cs.AIagentsDIP-CAD: A Design-Intent-Preserving Agent Harness for Parametric CAD Modeling
arXiv:2609.22688v3 Announce Type: replace-cross Abstract: Parametric CAD modeling requires not only matching the target shape, but also preserving design intent under parameter edits. However, most existing methods focus…
- arXiv cs.AIagentsFrom Code to Requirements: Agentic Reverse Engineering of Business Rules at Enterprise Scale
arXiv:2609.22719v1 Announce Type: cross Abstract: Business requirements for enterprise software systems are rarely captured in structured form; the logic resides instead in source code, configuration files, and instituti…
- arXiv cs.AIagentsMATE: Policy-Aware Security Auditing for Mobile Agents via Synthesis-Driven Trajectory Learning
arXiv:2609.22724v1 Announce Type: cross Abstract: Mobile agents powered by foundation models now automate complex, multi-step workflows on real devices, but their trajectories can violate app-specific security policies.…
- arXiv cs.AIagentsSelfOp: An Optimization Algorithm for Self-Improving Security Agents
arXiv:2609.22792v1 Announce Type: cross Abstract: LLM agents are increasingly used for security tasks: vulnerability discovery, exploit reproduction, and patch generation. Improving them at the model level demands expert…
- arXiv cs.AIagentsThe Price of Safety: Benign-Case Utility and Token Overhead of Memory-Poisoning Defenses in LLM Agents
arXiv:2609.22818v2 Announce Type: replace-cross Abstract: Memory-poisoning defenses for LLM agents are typically evaluated by how well they stop attacks, even though almost all traffic they encounter is benign. We measur…
- arXiv cs.AIagentsPlanning-Aligned Pretraining of BEV Representations with Sparse Action-Conditioned Targets for End-to-End Autonomous Driving
arXiv:2609.22868v1 Announce Type: cross Abstract: End-to-end driving requires planning-relevant bird's-eye-view (BEV) representations, but existing pretraining approaches often rely on task annotations or dense scene rec…
- arXiv cs.AIagentsTowards Full Pipeline FP8 Reinforcement Learning for LLMs
arXiv:2609.22870v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become a key technique for improving the reasoning and agentic abilities of large language models (LLMs). Although FP8 quantization can ac…
- arXiv cs.AIagentsThe Law of Stop: Interruptibility, Injunctions, and the Governance of Agentic AI
arXiv:2609.22882v2 Announce Type: replace-cross Abstract: On June 12, 2026, the U.S. government ordered Anthropic to bar foreign nationals from two of its most capable models. Unable to sort users by nationality, it with…
- arXiv cs.AIagentsAn Iterative LangGraph Agent for Text-to-SQL: Natural Language Access to the Chicago Crime Database
arXiv:2609.22917v2 Announce Type: replace-cross Abstract: Non-technical stakeholders frequently cannot write the SQL needed to extract insights from operational databases. We built and evaluated a Text-to-SQL agent that…
- arXiv cs.AIagentsAn Evolutionary Agentic Approach for Open-ended Image Quality Perception
arXiv:2609.22942v1 Announce Type: cross Abstract: Generative models are rapidly expanding image quality assessment (IQA) beyond traditional fidelity factors to emerging dimensions such as physical plausibility and text-r…
- arXiv cs.AIagentsNostrAgent: A Decentralized Identity and Delegation Architecture for Sovereign Agentic Systems
arXiv:2609.22944v1 Announce Type: cross Abstract: Autonomous AI agents increasingly act across organizational boundaries on behalf of human operators: they invoke third-party services, delegate subtasks to other agents,…
- arXiv cs.AIagentsBeyond Single-Model Injection: A Threat Model and Defense Architecture for Prompt Injection in Multi-Agent Systems
arXiv:2609.22949v1 Announce Type: cross Abstract: Existing prompt injection research focuses on single-model chatbot scenarios, where an attacker manipulates one LLM through crafted input. Multi-agent systems amplify thi…
- arXiv cs.AIagentsWhen Agentic Trust Crosses Organizational Boundaries: Structural Externalization and a Reference Model for Trust Evidence
arXiv:2609.22961v1 Announce Type: cross Abstract: Agentic systems increasingly invoke tools, services, data, and other agents across organizational boundaries, yet a relying party cannot assess a delegated action solely…