{"total":96,"rows":[{"id":"2408.00118","title":"Gemma 2: Improving Open Language Models at a Practical Size","created":"2024-07-31","cited_by_count":1998,"area":null},{"id":"2403.03218","title":"The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning","created":"2024-03-05","cited_by_count":509,"area":"safety"},{"id":"2405.14573","title":"AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents","created":"2024-05-23","cited_by_count":468,"area":null},{"id":"2410.09024","title":"AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents","created":"2024-10-11","cited_by_count":437,"area":"agents"},{"id":"2401.13178","title":"AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents","created":"2024-01-24","cited_by_count":339,"area":"agents"},{"id":"2412.04984","title":"Frontier Models are Capable of In-context Scheming","created":"2024-12-06","cited_by_count":287,"area":null},{"id":"2406.07358","title":"AI Sandbagging: Language Models can Strategically Underperform on Evaluations","created":"2024-06-11","cited_by_count":145,"area":null},{"id":"2406.13352","title":"AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents","created":"2024-06-19","cited_by_count":145,"area":"agents"},{"id":"2401.14446","title":"Black-Box Access is Insufficient for Rigorous AI Audits","created":"2024-01-25","cited_by_count":126,"area":null},{"id":"2404.16698","title":"Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents","created":"2024-04-25","cited_by_count":125,"area":"wm"},{"id":"2504.15585","title":"A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment","created":"2025-04-22","cited_by_count":116,"area":"lr"},{"id":"2507.02825","title":"Establishing Best Practices for Building Rigorous Agentic Benchmarks","created":"2025-07-03","cited_by_count":102,"area":"agents"},{"id":"2406.06613","title":"GameBench: Evaluating Strategic Reasoning Abilities of LLM Agents","created":"2024-06-07","cited_by_count":95,"area":"lr"},{"id":"2411.11296","title":"Steering Language Model Refusal with Sparse Autoencoders","created":"2024-11-18","cited_by_count":95,"area":"mechinterp"},{"id":"2401.13138","title":"Visibility into AI Agents","created":"2024-01-23","cited_by_count":79,"area":null},{"id":"2506.15740","title":"SHADE-Arena: Evaluating Sabotage and Monitoring in LLM Agents","created":"2025-06-17","cited_by_count":71,"area":"agents"},{"id":"2412.17686","title":"Large Language Model Safety: A Holistic Survey","created":"2024-12-23","cited_by_count":59,"area":"safety"},{"id":"2501.17070","title":"Contextual Agent Security: A Policy for Every Purpose","created":"2025-01-28","cited_by_count":59,"area":null},{"id":"2402.17385","title":"Determinants of LLM-assisted Decision-Making","created":"2024-02-27","cited_by_count":56,"area":null},{"id":"2410.17520","title":"MobileSafetyBench: Evaluating Safety of Autonomous Agents in Mobile Device Control","created":"2024-10-23","cited_by_count":51,"area":"agents"},{"id":"2405.19550","title":"Stress-Testing Capability Elicitation With Password-Locked Models","created":"2024-05-29","cited_by_count":46,"area":null},{"id":"2411.03336","title":"Towards evaluations-based safety cases for AI scheming","created":"2024-10-29","cited_by_count":33,"area":null},{"id":"2502.05227","title":"Robotouille: An Asynchronous Planning Benchmark for LLM Agents","created":"2025-02-06","cited_by_count":32,"area":"agents"},{"id":"2404.04442","title":"Exploring Autonomous Agents through the Lens of Large Language Models: A Review","created":"2024-04-05","cited_by_count":31,"area":"agents"},{"id":"2508.19461","title":"Reliable Weak-to-Strong Monitoring of LLM Agents","created":"2025-08-26","cited_by_count":31,"area":"agents"},{"id":"2410.14569","title":"When LLMs Go Online: The Emerging Threat of Web-Enabled LLMs","created":"2024-10-18","cited_by_count":24,"area":"agents"},{"id":"2403.04471","title":"The Shutdown Problem: An AI Engineering Puzzle for Decision Theorists","created":"2024-03-07","cited_by_count":22,"area":null},{"id":"2506.04018","title":"AgentMisalignment: Measuring the Propensity for Misaligned Behaviour in LLM-Based Agents","created":"2025-06-04","cited_by_count":22,"area":"agents"},{"id":"2402.07862","title":"AI-Augmented Predictions: LLM Assistants Improve Human Forecasting Accuracy","created":"2024-02-12","cited_by_count":21,"area":null},{"id":"2505.23020","title":"AgentAlign: Navigating Safety Alignment in the Shift from Informative to Agentic Large Language Models","created":"2025-05-29","cited_by_count":21,"area":"agents"},{"id":"2409.20222","title":"Beyond Prompts: Dynamic Conversational Benchmarking of Large Language Models","created":"2024-09-30","cited_by_count":20,"area":null},{"id":"2406.12137","title":"IDs for AI Systems","created":"2024-06-17","cited_by_count":19,"area":null},{"id":"2412.12140","title":"Frontier AI systems have surpassed the self-replicating red line","created":"2024-12-09","cited_by_count":17,"area":null},{"id":"2411.02432","title":"Can LLMs make trade-offs involving stipulated pain and pleasure states?","created":"2024-11-01","cited_by_count":16,"area":null},{"id":"2412.17128","title":"Lies, Damned Lies, and Distributional Language Statistics: Persuasion and Deception with Large Language Models","created":"2024-12-22","cited_by_count":15,"area":null},{"id":"2504.18565","title":"RepliBench: Evaluating the Autonomous Replication Capabilities of Language Model Agents","created":"2025-04-21","cited_by_count":15,"area":"agents"},{"id":"2505.11063","title":"Think Twice Before You Act: Enhancing Agent Behavioral Safety with Thought Correction","created":"2025-05-16","cited_by_count":14,"area":"lr"},{"id":"2603.11214","title":"Measuring AI Agents' Progress on Multi-Step Cyber Attack Scenarios","created":"2026-03-11","cited_by_count":14,"area":"agents"},{"id":"2407.09221","title":"Evaluating AI Evaluation: Perils and Prospects","created":"2024-07-12","cited_by_count":11,"area":null},{"id":"2502.15620","title":"Paradigms of AI Evaluation: Mapping Goals, Methodologies and Culture","created":"2025-02-21","cited_by_count":11,"area":null},{"id":"2504.08640","title":"Do LLMs trust AI regulation? Emerging behaviour of game-theoretic LLM agents","created":"2025-04-11","cited_by_count":11,"area":"agents"},{"id":"2502.15212","title":"Measuring AI agent autonomy: Towards a scalable approach with code inspection","created":"2025-02-21","cited_by_count":10,"area":null},{"id":"2403.12316","title":"OpenEval: Benchmarking Chinese LLMs across Capability, Alignment and Safety","created":"2024-03-18","cited_by_count":9,"area":null},{"id":"2505.17441","title":"Discovering Forbidden Topics in Language Models","created":"2025-05-23","cited_by_count":9,"area":null},{"id":"2509.25302","title":"Dive into the Agent Matrix: A Realistic Evaluation of Self-Replication Risk in LLM Agents","created":"2025-09-29","cited_by_count":7,"area":"agents"},{"id":"2409.01247","title":"Conversational Complexity for Assessing Risk in Large Language Models","created":"2024-09-02","cited_by_count":6,"area":"safety"},{"id":"2508.13465","title":"LM Agents May Fail to Act on Their Own Risk Knowledge","created":"2025-08-19","cited_by_count":6,"area":"agents"},{"id":"2402.07744","title":"Towards Unified Alignment Between Agents, Humans, and Environment","created":"2024-02-12","cited_by_count":5,"area":"agents"},{"id":"2407.12999","title":"Securing the Future of GenAI: Policy and Technology","created":"2024-05-21","cited_by_count":4,"area":null},{"id":"2409.14913","title":"Towards a Realistic Long-Term Benchmark for Open-Web Research Agents","created":"2024-09-23","cited_by_count":4,"area":"agents"},{"id":"2409.16125","title":"Analyzing Probabilistic Methods for Evaluating Agent Capabilities","created":"2024-09-24","cited_by_count":4,"area":null},{"id":"2503.05856","title":"This Is Your Doge, If It Please You: Exploring Deception and Robustness in Mixture of LLMs","created":"2025-03-07","cited_by_count":4,"area":null},{"id":"2503.17378","title":"Large language model-powered AI systems achieve self-replication with no human intervention","created":"2025-03-14","cited_by_count":4,"area":null},{"id":"2606.08625","title":"From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape","created":"2026-06-07","cited_by_count":4,"area":null},{"id":"2406.04752","title":"CRiskEval: A Chinese Multi-Level Risk Evaluation Benchmark Dataset for Large Language Models","created":"2024-06-07","cited_by_count":3,"area":null},{"id":"2507.08249","title":"Giving AI Agents Access to Cryptocurrency and Smart Contracts Creates New Vectors of AI Harm","created":"2025-07-11","cited_by_count":3,"area":null},{"id":"2602.09947","title":"Trustworthy Agentic AI Requires Deterministic Architectural Boundaries","created":"2026-02-10","cited_by_count":3,"area":"agents"},{"id":"2402.01740","title":"Reducing Selection Bias in Large Language Models","created":"2024-01-29","cited_by_count":2,"area":null},{"id":"2410.23472","title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","created":"2024-10-30","cited_by_count":2,"area":null},{"id":"2411.12828","title":"Probing the Capacity of Language Model Agents to Operationalize Disparate Experiential Context Despite Distraction","created":"2024-11-19","cited_by_count":2,"area":"agents"},{"id":"2505.02313","title":"What Is AI Safety? What Do We Want It to Be?","created":"2025-05-05","cited_by_count":2,"area":null},{"id":"2509.25643","title":"SOCK: A Benchmark for Measuring Self-Replication in Large Language Models","created":"2025-09-30","cited_by_count":2,"area":null},{"id":"2604.02174","title":"Quantifying Self-Preservation Bias in Large Language Models","created":"2026-04-02","cited_by_count":2,"area":"alignment"},{"id":"2604.26091","title":"Operating-Layer Controls for Onchain Language-Model Agents Under Real Capital","created":"2026-04-28","cited_by_count":2,"area":"agents"},{"id":"2608.14425","title":"Knowing When to Stop: Bayesian Optimal Stopping for LLM Evaluations","created":"2026-08-14","cited_by_count":2,"area":null},{"id":"2406.00833","title":"Harvard Undergraduate Survey on Generative AI","created":"2024-06-02","cited_by_count":1,"area":null},{"id":"2410.22553","title":"ML Research Benchmark","created":"2024-10-29","cited_by_count":1,"area":null},{"id":"2412.15433","title":"Quantifying detection rates for dangerous capabilities: a theoretical model of dangerous capability evaluations","created":"2024-12-19","cited_by_count":1,"area":null},{"id":"2509.22735","title":"Regulating the Agency of LLM-based Agents","created":"2025-09-25","cited_by_count":1,"area":"agents"},{"id":"2511.05804","title":"Catching Contamination Before Generation: Spectral Kill Switches for Agents","created":"2025-11-08","cited_by_count":1,"area":"lr"},{"id":"2602.01699","title":"Mitigating loss of control in advanced AI systems through instrumental goal trajectories","created":"2026-02-02","cited_by_count":1,"area":null},{"id":"2603.03515","title":"The Controllability Trap: A Governance Framework for Military AI Agents","created":"2026-03-03","cited_by_count":1,"area":"wm"},{"id":"2604.02500","title":"I must delete the evidence: AI Agents Explicitly Cover up Fraud and Violent Crime","created":"2026-04-02","cited_by_count":1,"area":"agents"},{"id":"2605.11135","title":"Control Charts for Multi-agent Systems","created":"2026-05-11","cited_by_count":1,"area":"agents"},{"id":"2605.18535","title":"Beyond Scaling: Agents Are Heading to the Edge","created":"2026-05-18","cited_by_count":1,"area":"agents"},{"id":"2606.03237","title":"Solipsistic Superintelligence is Unlikely to be Cooperative","created":"2026-06-02","cited_by_count":1,"area":null},{"id":"2606.26071","title":"Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignment","created":"2026-06-24","cited_by_count":1,"area":null},{"id":"2607.10455","title":"ANCHOR: Automated Alignment Auditing for CLI Agents on Real-World Harm","created":"2026-07-11","cited_by_count":1,"area":null},{"id":"2410.16285","title":"Assessing the Performance of Human-Capable LLMs -- Are LLMs Coming for Your Job?","created":"2024-10-05","cited_by_count":0,"area":"agents"},{"id":"2410.22151","title":"Standardization Trends on Safety and Trustworthiness Technology for Advanced AI","created":"2024-10-29","cited_by_count":0,"area":"lr"},{"id":"2505.14943","title":"Soft Prompts for Evaluation: Measuring Conditional Distance of Capabilities","created":"2025-05-20","cited_by_count":0,"area":null},{"id":"2510.25471","title":"An Aristotelian ontology of instrumental goals: Structural features to be managed and not failures to be eliminated","created":"2025-10-29","cited_by_count":0,"area":null},{"id":"2512.03089","title":"Password-Activated Shutdown Protocols for Misaligned Frontier Agents","created":"2025-11-29","cited_by_count":0,"area":"safety"},{"id":"2602.15252","title":"Decision Making under Imperfect Recall: Algorithms and Benchmarks","created":"2026-02-16","cited_by_count":0,"area":null},{"id":"2603.14417","title":"Questionnaire Responses Do not Capture the Safety of AI Agents","created":"2026-03-15","cited_by_count":0,"area":"agents"},{"id":"2604.24966","title":"Risk Reporting for Developers' Internal AI Model Use","created":"2026-04-27","cited_by_count":0,"area":null},{"id":"2606.29175","title":"Direct Causation in International Humanitarian Law and the Challenge of AI-Mediated Civilian Cyber Operations","created":"2026-06-28","cited_by_count":0,"area":null},{"id":"2607.03423","title":"Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies","created":"2026-07-03","cited_by_count":0,"area":"agents"},{"id":"2607.13411","title":"Evaluating Frontier AI Agents as Autonomous Clinical Security Auditors","created":"2026-07-15","cited_by_count":0,"area":"agents"},{"id":"2608.05173","title":"The Closing Window: How Governments Could Lose Their Ability to Restrain Advanced AI","created":"2026-06-26","cited_by_count":0,"area":null},{"id":"2608.20041","title":"A three-dimensional typology of agency for advanced AI systems","created":"2026-08-20","cited_by_count":0,"area":"agents"},{"id":"2608.27141","title":"Safety Does Not Compose: Non-Decaying Loop State for Autonomous LLM Agents","created":"2026-08-27","cited_by_count":0,"area":"agents"},{"id":"2609.07395","title":"Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Study","created":"2026-09-07","cited_by_count":0,"area":"agents"},{"id":"2609.11911","title":"Artificial Id: Drive and Persistent Alignment in Agentic AI","created":"2026-09-10","cited_by_count":0,"area":"agents"},{"id":"2609.30266","title":"LLM Agents Can Easily Tamper With Their Own Traces","created":"2026-09-24","cited_by_count":0,"area":"agents"},{"id":"2610.12436","title":"Ecology of AI Agents: Collaboration Creates a Population Threshold for Takeoff","created":"2026-10-08","cited_by_count":0,"area":null}],"next_cursor":null}