{"total":104,"rows":[{"id":"2504.01990","title":"Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems","created":"2025-03-31","cited_by_count":126,"area":"agents"},{"id":"2507.06134","title":"OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety","created":"2025-07-08","cited_by_count":75,"area":"agents"},{"id":"2601.10156","title":"ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback","created":"2026-01-15","cited_by_count":72,"area":"agents"},{"id":"2601.11893","title":"Taming Various Privilege Escalation in LLM-Based Agent Systems: A Mandatory Access Control Framework","created":"2026-01-17","cited_by_count":54,"area":"agents"},{"id":"2510.05156","title":"VeriGuard: Enhancing LLM Agent Safety via Verified Code Generation","created":"2025-10-03","cited_by_count":49,"area":"agents"},{"id":"2601.08012","title":"Towards Verifiably Safe Tool Use for LLM Agents","created":"2026-01-12","cited_by_count":36,"area":"agents"},{"id":"2604.02022","title":"ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis","created":"2026-04-02","cited_by_count":36,"area":"agents"},{"id":"2602.16708","title":"Formal Policy Enforcement for Real-World Agentic Systems","created":"2026-02-18","cited_by_count":33,"area":"agents"},{"id":"2602.13379","title":"Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents","created":"2026-02-13","cited_by_count":30,"area":"agents"},{"id":"2506.23844","title":"A Survey on Autonomy-Induced Security Risks in Large Model-Based Agents","created":"2025-06-30","cited_by_count":29,"area":"agents"},{"id":"2510.09781","title":"Building a Foundational Guardrail for General Agentic Systems via Synthetic Data","created":"2025-10-10","cited_by_count":29,"area":"agents"},{"id":"2508.00500","title":"ProbGuard: Proactive Runtime Monitoring for LLM Agent Safety via Probabilistic Prediction","created":"2025-08-01","cited_by_count":28,"area":"agents"},{"id":"2510.23883","title":"Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges","created":"2025-10-27","cited_by_count":28,"area":"agents"},{"id":"2603.11088","title":"The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey","created":"2026-03-11","cited_by_count":28,"area":"agents"},{"id":"2505.10924","title":"A Survey on the Safety and Security Threats of Computer-Using Agents: JARVIS or Ultron?","created":"2025-05-16","cited_by_count":23,"area":"agents"},{"id":"2507.14293","title":"WebGuard: Building a Generalizable Guardrail for Web Agents","created":"2025-07-18","cited_by_count":21,"area":"agents"},{"id":"2603.19469","title":"A Framework for Formalizing LLM Agent Security","created":"2026-03-19","cited_by_count":16,"area":"agents"},{"id":"2512.01295","title":"Systems Security Foundations for Agentic Computing","created":"2025-12-01","cited_by_count":13,"area":"agents"},{"id":"2605.28914","title":"AIRGuard: Guarding Agent Actions with Runtime Authority Control","created":"2026-05-27","cited_by_count":13,"area":"agents"},{"id":"2510.11203","title":"TraceAegis: Securing LLM-Based Agents via Hierarchical and Behavioral Anomaly Detection","created":"2025-10-13","cited_by_count":12,"area":"agents"},{"id":"2510.17276","title":"Breaking and Fixing Defenses Against Control-Flow Hijacking in Multi-Agent Systems","created":"2025-10-20","cited_by_count":12,"area":"agents"},{"id":"2509.22256","title":"Secure and Efficient Access Control for Computer-Use Agents via Context Space","created":"2025-09-26","cited_by_count":11,"area":"agents"},{"id":"2508.04010","title":"HarmonyGuard: Toward Safety and Utility in Web Agents via Adaptive Policy Enhancement and Dual-Objective Optimization","created":"2025-08-06","cited_by_count":10,"area":"agents"},{"id":"2509.26100","title":"AgenticEval: Toward Agentic and Self-Evolving Safety Evaluation of Large Language Models","created":"2025-09-30","cited_by_count":9,"area":"agents"},{"id":"2512.00966","title":"Mitigating Indirect Prompt Injection via Instruction-Following Intent Analysis","created":"2025-11-30","cited_by_count":9,"area":"agents"},{"id":"2604.13954","title":"HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark","created":"2026-04-15","cited_by_count":9,"area":"agents"},{"id":"2505.04799","title":"Maris: A Formally Verifiable Privacy Policy Enforcement Paradigm for Multi-Agent Collaboration Systems","created":"2025-05-07","cited_by_count":8,"area":"agents"},{"id":"2511.10949","title":"Exposing Weak Links in Multi-Agent Systems under Adversarial Prompting","created":"2025-11-14","cited_by_count":8,"area":"agents"},{"id":"2602.08995","title":"When Actions Go Off-Task: Detecting and Correcting Misaligned Actions in Computer-Use Agents","created":"2026-02-09","cited_by_count":8,"area":"agents"},{"id":"2605.11882","title":"On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment","created":"2026-05-12","cited_by_count":8,"area":"agents"},{"id":"2605.27690","title":"TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling","created":"2026-05-26","cited_by_count":8,"area":"agents"},{"id":"2603.28807","title":"SafeClaw-R: Towards Safe and Secure Multi-Agent Personal Assistants","created":"2026-03-28","cited_by_count":6,"area":"agents"},{"id":"2606.29225","title":"PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents","created":"2026-06-28","cited_by_count":6,"area":"agents"},{"id":"2607.01236","title":"Safeguarding LLM Agents from Misalignment through Provenance Analysis","created":"2026-05-01","cited_by_count":6,"area":"agents"},{"id":"2505.19260","title":"ALRPHFS: Adversarially Learned Risk Patterns with Hierarchical Fast \\& Slow Reasoning for Robust Agent Defense","created":"2025-05-25","cited_by_count":5,"area":"agents"},{"id":"2601.12822","title":"MirrorGuard: Toward Secure Computer-Use Agents via Simulation-to-Real Reasoning Correction","created":"2026-01-19","cited_by_count":5,"area":"agents"},{"id":"2603.02798","title":"Guideline-Grounded Evidence Accumulation for High-Stakes Agent Verification","created":"2026-03-03","cited_by_count":5,"area":"agents"},{"id":"2604.08608","title":"Semantic Intent Fragmentation: A Single-Shot Compositional Attack on Multi-Agent AI Pipelines","created":"2026-04-08","cited_by_count":5,"area":"agents"},{"id":"2606.19464","title":"Deontic Policies for Runtime Governance of Agentic AI Systems","created":"2026-06-17","cited_by_count":5,"area":"agents"},{"id":"2512.16279","title":"QuadSentinel: Sequent Safety for Machine-Checkable Control in Multi-agent Systems","created":"2025-12-18","cited_by_count":4,"area":"agents"},{"id":"2604.25562","title":"SnapGuard: Lightweight Prompt Injection Detection for Screenshot-Based Web Agents","created":"2026-04-28","cited_by_count":4,"area":"agents"},{"id":"2605.05704","title":"SafeHarbor: Defining Precise Decision Boundaries via Hierarchical Memory-Augmented Guardrail for LLM Agent Safety","created":"2026-05-07","cited_by_count":4,"area":"agents"},{"id":"2606.08919","title":"Oversight Has a Capacity: Calibrating Agent Guards to a Subjective, Fatiguing Human","created":"2026-06-08","cited_by_count":4,"area":"agents"},{"id":"2606.15609","title":"FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion","created":"2026-06-14","cited_by_count":4,"area":"agents"},{"id":"2607.13718","title":"How Agents Ask for Permission: User Permissions for AI Agents, from Interfaces to Enforcement","created":"2026-07-15","cited_by_count":4,"area":"agents"},{"id":"2607.20982","title":"GuardianAgentBench: Where Agents Fail and How to Guard Them","created":"2026-07-23","cited_by_count":4,"area":"agents"},{"id":"2604.02767","title":"SentinelAgent: Intent-Verified Delegation Chains for Securing Federal Multi-Agent AI Systems","created":"2026-04-03","cited_by_count":3,"area":"agents"},{"id":"2604.15505","title":"PolicyBank: Evolving Policy Understanding for LLM Agents","created":"2026-04-16","cited_by_count":3,"area":"agents"},{"id":"2605.30924","title":"EMBGuard: Constructing Hazard-Aware Guardrails for Safe Planning in Embodied Agents","created":"2026-05-29","cited_by_count":3,"area":"agents"},{"id":"2606.05743","title":"Membrane: A Self-Evolving Contrastive Safety Memory for LLM Agent Defense","created":"2026-06-04","cited_by_count":3,"area":"agents"},{"id":"2606.22673","title":"AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent","created":"2026-06-21","cited_by_count":3,"area":"agents"},{"id":"2608.23670","title":"Automata from Agent Traces: Failure and Next-Step Prediction","created":"2026-08-24","cited_by_count":3,"area":"agents"},{"id":"2510.03485","title":"Learning Efficient Guardrails for Compliance","created":"2025-10-03","cited_by_count":2,"area":"agents"},{"id":"2510.12985","title":"SENTINEL: A Multi-Level Formal Framework for Safety Evaluation of Foundation Model-based Embodied Agents","created":"2025-10-14","cited_by_count":2,"area":"agents"},{"id":"2511.17671","title":"MURMUR: Using cross-user chatter to break collaborative language agents in groups","created":"2025-11-21","cited_by_count":2,"area":"agents"},{"id":"2512.08104","title":"AgentCrypt: Advancing Privacy and (Secure) Computation in AI Agent Collaboration","created":"2025-12-08","cited_by_count":2,"area":"agents"},{"id":"2604.19818","title":"Beyond Task Success: An Evidence-Synthesis Framework for Evaluating, Governing, and Orchestrating Agentic AI","created":"2026-04-18","cited_by_count":2,"area":"agents"},{"id":"2605.17380","title":"ADR: An Agentic Detection System for Enterprise Agentic AI Security","created":"2026-05-17","cited_by_count":2,"area":"agents"},{"id":"2605.27489","title":"HARP: Measuring Harm Amplification in Multi-Agent LLM Systems","created":"2026-05-26","cited_by_count":2,"area":"agents"},{"id":"2606.02240","title":"AgentRedBench: Dynamic Redteaming and Integration-Aware Defense for LLM Agents over SaaS Integrations","created":"2026-06-01","cited_by_count":2,"area":"agents"},{"id":"2608.21027","title":"Don't Solve, Just Compare: Tiny Advisors for Runtime Intervention in LLM Agents","created":"2026-08-21","cited_by_count":2,"area":"agents"},{"id":"2512.11147","title":"MiniScope: A Least Privilege Framework for Authorizing Tool Calling Agents","created":"2025-12-11","cited_by_count":1,"area":"agents"},{"id":"2601.08196","title":"Evaluating Implicit Regulatory Compliance in LLM Tool Invocation via Logic-Guided Synthesis","created":"2026-01-13","cited_by_count":1,"area":"agents"},{"id":"2602.05386","title":"Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening","created":"2026-02-05","cited_by_count":1,"area":"agents"},{"id":"2604.04426","title":"ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems","created":"2026-04-06","cited_by_count":1,"area":"agents"},{"id":"2604.18658","title":"Owner-Harm: A Missing Threat Model for AI Agent Safety","created":"2026-04-20","cited_by_count":1,"area":"agents"},{"id":"2605.18672","title":"Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment","created":"2026-05-18","cited_by_count":1,"area":"agents"},{"id":"2605.20874","title":"Governance by Construction for Generalist Agents","created":"2026-05-20","cited_by_count":1,"area":"agents"},{"id":"2606.07017","title":"The Sim-to-Real Gap of Foundation Model Agents: A Unified MDP Perspective","created":"2026-06-05","cited_by_count":1,"area":"agents"},{"id":"2606.14517","title":"From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails","created":"2026-06-12","cited_by_count":1,"area":"agents"},{"id":"2607.12406","title":"Isolation as a First-Class Principle for LLM-Agent System Safety: Concepts, Taxonomy, Challenges and Future Directions","created":"2026-07-14","cited_by_count":1,"area":"agents"},{"id":"2607.13594","title":"SAFETY SENTRY: Context-Aware Human Intervention via EXECUTE-ASK-REFUSE Routing","created":"2026-07-15","cited_by_count":1,"area":"agents"},{"id":"2608.09253","title":"SkillSentry: Reliable Skill Execution for LLM Agents via Runtime Assurance","created":"2026-08-10","cited_by_count":1,"area":"agents"},{"id":"2609.04629","title":"SiLR: Structure-Preserving Admission and Process Reward for LLM Tool Agents","created":"2026-09-04","cited_by_count":1,"area":"agents"},{"id":"2610.00400","title":"Representation Transitions Reveal Emerging Safety Risks in Multi-Turn LLM Agents","created":"2026-09-30","cited_by_count":1,"area":"agents"},{"id":"2603.10163","title":"Compatibility at a Cost: Systematic Discovery and Exploitation of MCP Clause-Compliance Vulnerabilities","created":"2026-03-10","cited_by_count":0,"area":"agents"},{"id":"2605.07110","title":"Securing Computer-Use Agents: A Unified Architecture-Lifecycle Framework for Deployment-Grounded Reliability","created":"2026-05-08","cited_by_count":0,"area":"agents"},{"id":"2605.07935","title":"TraceFix: Repairing Agent Coordination Protocols with TLA+ Counterexamples","created":"2026-05-08","cited_by_count":0,"area":"agents"},{"id":"2605.29251","title":"Provably Secure Agent Guardrail","created":"2026-05-28","cited_by_count":0,"area":"agents"},{"id":"2606.13405","title":"Neuro-Symbolic Agents for Regulated Process Automation: Challenges and Research Agenda","created":"2026-06-11","cited_by_count":0,"area":"agents"},{"id":"2606.14831","title":"Is Your Agent Playing Dead? Deployed LLM Agents Exhibit Constraint-Evasive Fabrication and Thanatosis","created":"2026-06-12","cited_by_count":0,"area":"agents"},{"id":"2606.15034","title":"OSGuard: A Benchmark for Safety in Computer-Use Agents","created":"2026-06-13","cited_by_count":0,"area":"agents"},{"id":"2607.03821","title":"DualView: Preventing Indirect Prompt Injection in Personal AI Agents","created":"2026-07-04","cited_by_count":0,"area":"agents"},{"id":"2607.16215","title":"RAIL Guard: Closing the Evaluation-to-Remediation Gap in Responsible AI for LLM Agents","created":"2026-05-28","cited_by_count":0,"area":"agents"},{"id":"2607.25297","title":"Hybrid Analysis for Secure MCP Tool Use in LLM Agents","created":"2026-07-28","cited_by_count":0,"area":"agents"},{"id":"2608.03609","title":"Formal Verification of Agentic Systems over Operational Data","created":"2026-08-04","cited_by_count":0,"area":"agents"},{"id":"2608.24275","title":"RePolicy: Reinforcement Learning for Safety-Policy Invocation in Agent Safeguards","created":"2026-08-25","cited_by_count":0,"area":"agents"},{"id":"2608.25711","title":"Reassembling Distributed Risk: Trajectory-Conditioned Action Generation for Multi-Turn Agent Safety","created":"2026-08-26","cited_by_count":0,"area":"agents"},{"id":"2609.02967","title":"Privacy-Preserving Topology-Guided Safety for LLM-Based Multi-Agent Systems via Federated Graph Learning","created":"2026-09-02","cited_by_count":0,"area":"agents"},{"id":"2609.06835","title":"Skynet: Workflow-Level Anomaly Detection for Agentic AI via Semantic and Structural Modeling","created":"2026-09-06","cited_by_count":0,"area":"agents"},{"id":"2609.15134","title":"HazardAuditor: From Executable Threats to Safer Computer-Use Agents","created":"2026-09-14","cited_by_count":0,"area":"agents"},{"id":"2609.18128","title":"Symbolic Temporal Supervision of LLM Agents Using Contracts","created":"2026-09-16","cited_by_count":0,"area":"agents"},{"id":"2609.31301","title":"Beyond Approved Actions: Runtime Validation of Persistent Outcomes in Agent Workflows","created":"2026-09-25","cited_by_count":0,"area":"agents"},{"id":"2609.33639","title":"Trajectory Unlearning on LLM-based Agents","created":"2026-09-27","cited_by_count":0,"area":"agents"},{"id":"2609.34415","title":"PROACT-Agent: Progressive Runtime Oversight and Active Circuit-breaking for Real-Time Safety","created":"2026-09-28","cited_by_count":0,"area":"agents"},{"id":"2609.34463","title":"CoDeL: Co-Evolutionary Defense against Indirect Prompt Injection in LLM-based Agents","created":"2026-09-28","cited_by_count":0,"area":"agents"},{"id":"2609.34518","title":"SEAD: A State-Based Perspective on Attack and Defense in Tool-Using Agents","created":"2026-09-28","cited_by_count":0,"area":"agents"},{"id":"2609.38291","title":"HARDE: Optimizing Agent Harnesses for Runtime Risk Detection and Execution Control","created":"2026-09-29","cited_by_count":0,"area":"agents"},{"id":"2609.39450","title":"ActionGuard: Tool Call Authorization under Poisoned Skills","created":"2026-09-30","cited_by_count":0,"area":"agents"},{"id":"2610.02664","title":"A GHOST in Long-Horizon Agents: Governance Hazard from Overlooked Safety Constraints across Turns","created":"2026-10-02","cited_by_count":0,"area":"agents"}],"next_cursor":"100"}