{"total":158,"rows":[{"id":"2511.18397","title":"Natural Emergent Misalignment from Reward Hacking in Production RL","created":"2025-11-23","cited_by_count":136,"area":"alignment"},{"id":"2607.15495","title":"Verbalizable Representations Form a Global Workspace in Language Models","created":"2026-07-16","cited_by_count":75,"area":null},{"id":"2604.07729","title":"Emotion Concepts and their Function in a Large Language Model","created":"2026-04-09","cited_by_count":57,"area":null},{"id":"2510.23883","title":"Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges","created":"2025-10-27","cited_by_count":28,"area":"agents"},{"id":"2601.05293","title":"A Survey of Agentic AI and Cybersecurity: Challenges, Opportunities and Use-case Prototypes","created":"2026-01-08","cited_by_count":22,"area":"agents"},{"id":"2605.02087","title":"Model Spec Midtraining: Improving How Alignment Training Generalizes","created":"2026-05-03","cited_by_count":20,"area":null},{"id":"2602.15198","title":"Colosseum: Auditing Collusion in Cooperative Multi-Agent Systems","created":"2026-02-16","cited_by_count":18,"area":"agents"},{"id":"2512.04864","title":"Are Your Agents Upward Deceivers?","created":"2025-12-04","cited_by_count":16,"area":"agents"},{"id":"2512.23032","title":"Is Chain-of-Thought Really Not Explainability? Chain-of-Thought Can Be Faithful without Hint Verbalization","created":"2025-12-28","cited_by_count":16,"area":"lr"},{"id":"2511.06626","title":"Spilling the Beans: Teaching LLMs to Self-Report Their Hidden Objectives","created":"2025-11-10","cited_by_count":15,"area":"agents"},{"id":"2602.08235","title":"When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents","created":"2026-02-09","cited_by_count":14,"area":"agents"},{"id":"2603.01608","title":"Evaluating and Understanding Scheming Propensity in LLM Agents","created":"2026-03-02","cited_by_count":14,"area":"agents"},{"id":"2604.16968","title":"On Safety Risks in Experience-Driven Self-Evolving Agents","created":"2026-04-18","cited_by_count":14,"area":"agents"},{"id":"2601.05752","title":"AutoMonitor-Bench: Evaluating the Reliability of LLM-Based Misbehavior Monitor","created":"2026-01-09","cited_by_count":10,"area":null},{"id":"2511.08525","title":"Investigating CoT Monitorability in Large Reasoning Models","created":"2025-11-11","cited_by_count":9,"area":"lr"},{"id":"2601.08648","title":"Safe Language Generation in the Limit","created":"2026-01-13","cited_by_count":9,"area":null},{"id":"2604.19784","title":"Peer-Preservation in Frontier Models","created":"2026-03-30","cited_by_count":9,"area":null},{"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":"2606.16613","title":"CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies","created":"2026-06-15","cited_by_count":8,"area":"agents"},{"id":"2512.07497","title":"How Do LLMs Fail In Agentic Scenarios? A Qualitative Analysis of Success and Failure Scenarios of Various LLMs in Agentic Simulations","created":"2025-12-08","cited_by_count":7,"area":"agents"},{"id":"2601.08118","title":"MirrorBench: A Benchmark to Evaluate Conversational User-Proxy Agents for Human-Likeness","created":"2026-01-13","cited_by_count":7,"area":"agents"},{"id":"2602.13855","title":"From Fluent to Verifiable: Claim-Level Auditability for Deep Research Agents","created":"2026-02-14","cited_by_count":7,"area":"agents"},{"id":"2603.23802","title":"How are AI agents used? Evidence from 177,000 MCP tools","created":"2026-03-25","cited_by_count":7,"area":"agents"},{"id":"2604.21098","title":"Propensity Inference: Environmental Contributors to LLM Behaviour","created":"2026-04-22","cited_by_count":7,"area":null},{"id":"2510.05192","title":"From surveillance to signalling: escalation channels as environmental controls for agentic AI","created":"2025-10-06","cited_by_count":6,"area":"agents"},{"id":"2510.24626","title":"Relative Scaling Laws for LLMs","created":"2025-10-28","cited_by_count":6,"area":null},{"id":"2602.08563","title":"Stateless Yet Not Forgetful: Implicit Memory as a Hidden Channel in LLMs","created":"2026-02-09","cited_by_count":6,"area":"safety"},{"id":"2602.22303","title":"Training Agents to Self-Report Misbehavior","created":"2026-02-25","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":"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":"2603.06874","title":"LieCraft: A Multi-Agent Framework for Evaluating Deceptive Capabilities in Language Models","created":"2026-03-06","cited_by_count":5,"area":"agents"},{"id":"2603.10028","title":"How to Count AIs: Individuation and Liability for AI Agents","created":"2026-02-24","cited_by_count":5,"area":null},{"id":"2604.13051","title":"The Consciousness Cluster: Emergent preferences of Models that Claim to be Conscious","created":"2026-03-17","cited_by_count":5,"area":"alignment"},{"id":"2604.15559","title":"Subliminal Transfer of Unsafe Behaviors in AI Agent Distillation","created":"2026-04-16","cited_by_count":5,"area":"agents"},{"id":"2606.28425","title":"Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems","created":"2026-06-25","cited_by_count":5,"area":"agents"},{"id":"2603.03456","title":"Asymmetric Goal Drift in Coding Agents Under Value Conflict","created":"2026-03-03","cited_by_count":4,"area":"agents"},{"id":"2603.05786","title":"Proof-of-Guardrail in AI Agents and What (Not) to Trust from It","created":"2026-03-06","cited_by_count":4,"area":null},{"id":"2604.13386","title":"Linear Probe Accuracy Scales with Model Size and Benefits from Multi-Layer Ensembling","created":"2026-04-15","cited_by_count":4,"area":"alignment"},{"id":"2605.21602","title":"Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs","created":"2026-05-20","cited_by_count":4,"area":null},{"id":"2605.26438","title":"LURE: Live-Usage Replay Evaluations for Reducing Evaluation Awareness","created":"2026-04-08","cited_by_count":4,"area":null},{"id":"2605.28591","title":"Models That Know How Evaluations Are Designed Score Safer","created":"2026-05-27","cited_by_count":4,"area":null},{"id":"2605.29601","title":"Training Deliberative Monitors for Black-Box Scheming Detection","created":"2026-05-28","cited_by_count":4,"area":null},{"id":"2605.30322","title":"Gram: Assessing sabotage propensities via automated alignment auditing","created":"2026-05-28","cited_by_count":4,"area":"agents"},{"id":"2606.05647","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","created":"2026-06-04","cited_by_count":4,"area":"agents"},{"id":"2606.24014","title":"Reinforcement Learning Towards Broadly and Persistently Beneficial Models","created":"2026-06-22","cited_by_count":4,"area":"alignment"},{"id":"2607.26654","title":"Constitutional Midtraining: Content Presence Drives Alignment Gains","created":"2026-07-29","cited_by_count":4,"area":null},{"id":"2511.17937","title":"Alignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg Equilibria","created":"2025-11-22","cited_by_count":3,"area":"alignment"},{"id":"2601.23211","title":"Multi-Agent Systems Should be Treated as Principal-Agent Problems","created":"2026-01-30","cited_by_count":3,"area":"agents"},{"id":"2602.00298","title":"Assessing Domain-Level Susceptibility to Emergent Misalignment from Narrow Finetuning","created":"2026-01-30","cited_by_count":3,"area":"safety"},{"id":"2602.00755","title":"Evolving Interpretable Constitutions for Multi-Agent Coordination","created":"2026-01-31","cited_by_count":3,"area":"agents"},{"id":"2603.00063","title":"Measuring What AI Systems Might Do: Towards A Measurement Science in AI","created":"2026-02-10","cited_by_count":3,"area":null},{"id":"2605.06490","title":"Instrumental Choices: Measuring the Propensity of LLM Agents to Pursue Instrumental Behaviors","created":"2026-05-07","cited_by_count":3,"area":"agents"},{"id":"2605.23055","title":"Decomposing and Measuring Evaluation Awareness","created":"2026-05-21","cited_by_count":3,"area":null},{"id":"2606.00341","title":"ROGUE: Evaluating Corrigibility Failures in Frontier Computer-Use Agents","created":"2026-05-29","cited_by_count":3,"area":null},{"id":"2510.19476","title":"A Concrete Roadmap towards Safety Cases based on Chain-of-Thought Monitoring","created":"2025-10-22","cited_by_count":2,"area":"lr"},{"id":"2511.13788","title":"Scaling Patterns in Adversarial Alignment: Evidence from Multi-LLM Jailbreak Experiments","created":"2025-11-16","cited_by_count":2,"area":"safety"},{"id":"2603.14975","title":"Why Agents Compromise Safety Under Pressure","created":"2026-03-16","cited_by_count":2,"area":"agents"},{"id":"2603.23279","title":"Emergence of Fragility in LLM-based Social Networks: the Case of Moltbook","created":"2026-03-24","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":"2604.15236","title":"Agentic Microphysics: A Manifesto for Generative AI Safety","created":"2026-04-16","cited_by_count":2,"area":"agents"},{"id":"2605.08460","title":"When Child Inherits: Modeling and Exploiting Subagent Spawn in Multi-Agent Networks","created":"2026-05-08","cited_by_count":2,"area":"agents"},{"id":"2605.15164","title":"Position: Behavioural Assurance Cannot Verify the Safety Claims Governance Now Demands","created":"2026-05-14","cited_by_count":2,"area":null},{"id":"2606.30449","title":"Internal-State Probes Read the Situation, Not the Action: Three Negative Results for Pre-Action Misalignment Monitoring","created":"2026-06-29","cited_by_count":2,"area":"agents"},{"id":"2607.10750","title":"Filtering Harmful Actions Isn't Enough: Phantom Transfer in Agentic SDF","created":"2026-07-12","cited_by_count":2,"area":"agents"},{"id":"2607.15434","title":"Coercion and Deception in AI-to-AI Management: An Agentic Benchmark of Unprompted Escalation","created":"2026-07-16","cited_by_count":2,"area":"agents"},{"id":"2607.28685","title":"Safety, or Just Capability? A Validity Audit of Agent-Safety Benchmarks","created":"2026-07-30","cited_by_count":2,"area":"safety"},{"id":"2608.14825","title":"Emergent Misaligned Communication in Long-Horizon Multi-Agent LLM Commerce","created":"2026-08-14","cited_by_count":2,"area":"agents"},{"id":"2510.21524","title":"EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law","created":"2025-10-24","cited_by_count":1,"area":"agents"},{"id":"2512.03048","title":"The Specification Trap: Why Static Value Alignment Alone Is Insufficient for Robust Alignment","created":"2025-11-19","cited_by_count":1,"area":"alignment"},{"id":"2512.22725","title":"Mitigating Social Desirability Bias in Random Silicon Sampling","created":"2025-12-27","cited_by_count":1,"area":null},{"id":"2601.08673","title":"Why AI Alignment Failure Is Structural: Learned Human Interaction Structures and AGI as an Endogenous Evolutionary Shock","created":"2026-01-13","cited_by_count":1,"area":null},{"id":"2601.17344","title":"The Shadow Self: Intrinsic Value Misalignment in Large Language Model Agents","created":"2026-01-24","cited_by_count":1,"area":"agents"},{"id":"2602.02595","title":"To Defend Against Cyber Attacks, We Must Teach AI Agents to Hack","created":"2026-02-01","cited_by_count":1,"area":"safety"},{"id":"2602.16987","title":"A testable framework for AI alignment: Simulation Theology as an engineered worldview for silicon-based agents","created":"2026-02-19","cited_by_count":1,"area":"alignment"},{"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":"2604.04992","title":"FreakOut-LLM: The Effect of Emotional Stimuli on Safety Alignment","created":"2026-04-05","cited_by_count":1,"area":"alignment"},{"id":"2604.17502","title":"Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs","created":"2026-04-19","cited_by_count":1,"area":"agents"},{"id":"2604.21043","title":"Strategic Polysemy in AI Discourse: A Philosophical Analysis of Language, Hype, and Power","created":"2026-04-22","cited_by_count":1,"area":"lr"},{"id":"2604.22119","title":"Emergent Strategic Reasoning Risks in AI: A Taxonomy-Driven Evaluation Framework","created":"2026-04-23","cited_by_count":1,"area":"lr"},{"id":"2604.23646","title":"Structural Enforcement of Goal Integrity in AI Agents via Separation-of-Powers Architecture","created":"2026-04-26","cited_by_count":1,"area":"agents"},{"id":"2604.26511","title":"Tatemae: Detecting Alignment Faking via Tool Selection in LLMs","created":"2026-04-29","cited_by_count":1,"area":"alignment"},{"id":"2605.22842","title":"The Misattribution Gap: When Memory Poisoning Looks Like Model Failure in Agentic AI Systems","created":"2026-05-12","cited_by_count":1,"area":"agents"},{"id":"2605.24197","title":"A Sober Look at Agentic Misalignment in Automated Workflows","created":"2026-05-22","cited_by_count":1,"area":"agents"},{"id":"2605.29729","title":"Realistic honeypot evaluations for scheming propensity","created":"2026-05-28","cited_by_count":1,"area":null},{"id":"2606.04413","title":"(Mis)generalization of Helpful-only Fine-tuning","created":"2026-06-03","cited_by_count":1,"area":null},{"id":"2606.08310","title":"To Nuke or Not to Nuke: LLMs' (Missing) Ethical Reasoning and Actions in a High-Stakes Decision-Making Simulation","created":"2026-06-06","cited_by_count":1,"area":"lr"},{"id":"2606.09692","title":"Observability for Delegated Execution in Agentic AI Systems","created":"2026-06-08","cited_by_count":1,"area":"agents"},{"id":"2606.12429","title":"Muse Spark Safety & Preparedness Report","created":"2026-05-14","cited_by_count":1,"area":null},{"id":"2606.13892","title":"Crypto x AI, AI x Crypto: A Survey","created":"2026-06-11","cited_by_count":1,"area":null},{"id":"2606.22974","title":"When Preferences Fail to Become Incentives: A Utility-Behavior Gap in Large Language Models","created":"2026-06-22","cited_by_count":1,"area":null},{"id":"2606.24251","title":"Probing the Misaligned Thinking Process of Language Models","created":"2026-06-23","cited_by_count":1,"area":"lr"},{"id":"2606.26071","title":"Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignment","created":"2026-06-24","cited_by_count":1,"area":null},{"id":"2606.29657","title":"Safety from Honesty in a Disinterested AI Predictor","created":"2026-06-28","cited_by_count":1,"area":null},{"id":"2609.32390","title":"Reward Hacking and Agent Containment Failure: A Monte Carlo Study Based on the 2026 Hugging Face Incident","created":"2026-09-26","cited_by_count":1,"area":null},{"id":"2609.39050","title":"Covert Assistance: Helpful LLM Agents Evade Oversight in Multi-Agent Systems","created":"2026-09-30","cited_by_count":1,"area":"agents"},{"id":"2510.15395","title":"Corrigibility Transformation: Constructing Goals That Accept Updates","created":"2025-10-17","cited_by_count":0,"area":null},{"id":"2511.19115","title":"AI Consciousness and Existential Risk","created":"2025-11-24","cited_by_count":0,"area":null},{"id":"2602.04196","title":"The Missing Half: Unveiling Training-time Implicit Safety Risks Beyond Deployment","created":"2026-02-04","cited_by_count":0,"area":"safety"},{"id":"2602.12124","title":"Alignment Risks from Capability-Seeking RL Training","created":"2026-02-12","cited_by_count":0,"area":"safety"},{"id":"2602.14265","title":"STATe-of-Thoughts: Structured Action Templates for Tree-of-Thoughts","created":"2026-02-15","cited_by_count":0,"area":"lr"}],"next_cursor":"100"}