{"total":126,"rows":[{"id":"2505.23836","title":"Large Language Models Often Know When They Are Being Evaluated","created":"2025-05-28","cited_by_count":128,"area":null},{"id":"2502.06559","title":"Can We Trust AI Benchmarks? An Interdisciplinary Review of Current Issues in AI Evaluation","created":"2025-02-10","cited_by_count":60,"area":null},{"id":"2505.01420","title":"Evaluating Frontier Models for Stealth and Situational Awareness","created":"2025-05-02","cited_by_count":37,"area":null},{"id":"2507.01786","title":"Probing and Steering Evaluation Awareness of Language Models","created":"2025-07-02","cited_by_count":35,"area":null},{"id":"2504.02922","title":"Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning","created":"2025-04-03","cited_by_count":34,"area":null},{"id":"2502.12206","title":"Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals?","created":"2025-02-16","cited_by_count":33,"area":null},{"id":"2505.18889","title":"Security Concerns for Large Language Models: A Survey","created":"2025-05-24","cited_by_count":29,"area":null},{"id":"2507.12691","title":"Benchmarking Deception Probes via Black-to-White Performance Boosts","created":"2025-07-16","cited_by_count":19,"area":null},{"id":"2604.17596","title":"Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories","created":"2026-04-19","cited_by_count":15,"area":null},{"id":"2511.22662","title":"Difficulties with Evaluating a Deception Detector for AIs","created":"2025-11-27","cited_by_count":13,"area":null},{"id":"2509.23023","title":"Deceive, Detect, and Disclose: Large Language Models Play Mini-Mafia","created":"2025-09-27","cited_by_count":9,"area":null},{"id":"2512.17639","title":"Linear Personality Probing and Steering in LLMs: A Big Five Study","created":"2025-12-19","cited_by_count":9,"area":null},{"id":"2603.03992","title":"Measuring AI R&D Automation","created":"2026-03-04","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":"2509.04373","title":"Measuring Bias or Measuring the Task: Understanding the Brittle Nature of LLM Gender Biases","created":"2025-09-04","cited_by_count":8,"area":null},{"id":"2509.20393","title":"The Secret Agenda: LLMs Strategically Lie and Our Current Safety Tools Are Blind","created":"2025-09-23","cited_by_count":8,"area":null},{"id":"2510.08211","title":"LLMs Deceive Unintentionally: Emergent Misalignment in Dishonesty from Misaligned Samples to Biased Human-AI Interactions","created":"2025-10-09","cited_by_count":8,"area":null},{"id":"2503.16861","title":"In-House Evaluation Is Not Enough: Towards Robust Third-Party Flaw Disclosure for General-Purpose AI","created":"2025-03-21","cited_by_count":7,"area":null},{"id":"2504.20084","title":"AI Awareness","created":"2025-04-25","cited_by_count":7,"area":null},{"id":"2510.03999","title":"LH-Deception: Simulating and Understanding LLM Deceptive Behaviors in Long-Horizon Interactions","created":"2025-10-05","cited_by_count":7,"area":null},{"id":"2604.21098","title":"Propensity Inference: Environmental Contributors to LLM Behaviour","created":"2026-04-22","cited_by_count":7,"area":null},{"id":"2606.07612","title":"Position: Anthropomorphic Misalignment Research Needs Stronger Evidence","created":"2026-05-29","cited_by_count":7,"area":null},{"id":"2508.14927","title":"AI Testing Should Account for Sophisticated Strategic Behaviour","created":"2025-08-19","cited_by_count":6,"area":null},{"id":"2510.00857","title":"ManagerBench: Evaluating the Safety-Pragmatism Trade-off in Autonomous LLMs","created":"2025-10-01","cited_by_count":6,"area":null},{"id":"2510.24222","title":"HACK: Hallucinations Along Certainty and Knowledge Axes","created":"2025-10-28","cited_by_count":6,"area":null},{"id":"2511.02823","title":"Optimizing AI Agent Attacks With Synthetic Data","created":"2025-11-04","cited_by_count":6,"area":null},{"id":"2603.03308","title":"Old Habits Die Hard: How Conversational History Geometrically Traps LLMs","created":"2026-02-08","cited_by_count":6,"area":null},{"id":"2604.12493","title":"Latent Planning Emerges with Scale","created":"2026-04-14","cited_by_count":6,"area":null},{"id":"2505.19806","title":"Exploring Consciousness in LLMs: A Systematic Survey of Theories, Implementations, and Frontier Risks","created":"2025-05-26","cited_by_count":5,"area":null},{"id":"2510.12826","title":"Scheming Ability in LLM-to-LLM Strategic Interactions","created":"2025-10-11","cited_by_count":5,"area":null},{"id":"2607.07368","title":"Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitors","created":"2026-07-08","cited_by_count":5,"area":null},{"id":"2503.01470","title":"Position: Ensuring mutual privacy is necessary for effective external evaluation of proprietary AI systems","created":"2025-03-03","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":"2506.03755","title":"Misalignment or misuse? The AGI alignment tradeoff","created":"2025-06-04","cited_by_count":4,"area":null},{"id":"2506.21584","title":"Empirical Evidence for Alignment Faking in a Small LLM and Prompt-Based Mitigation Techniques","created":"2025-06-17","cited_by_count":4,"area":null},{"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":"2605.29601","title":"Training Deliberative Monitors for Black-Box Scheming Detection","created":"2026-05-28","cited_by_count":4,"area":null},{"id":"2507.02977","title":"LLMs are Capable of Misaligned Behavior Under Explicit Prohibition and Surveillance","created":"2025-06-30","cited_by_count":3,"area":null},{"id":"2507.23330","title":"AI Must not be Fully Autonomous","created":"2025-07-31","cited_by_count":3,"area":null},{"id":"2508.13247","title":"Goal-Directedness is in the Eye of the Beholder","created":"2025-08-18","cited_by_count":3,"area":null},{"id":"2510.26518","title":"Human-AI Complementarity: A Goal for Amplified Oversight","created":"2025-10-30","cited_by_count":3,"area":null},{"id":"2604.09104","title":"Scheming in the wild: detecting real-world AI scheming incidents with open-source intelligence","created":"2026-04-10","cited_by_count":3,"area":null},{"id":"2604.20995","title":"Value-Conflict Diagnostics Reveal Widespread Alignment Faking in Language Models","created":"2026-04-22","cited_by_count":3,"area":null},{"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":"2607.14345","title":"Value Leakage: An LLM's Answers Are Silently Shaped by Its Own Values","created":"2026-07-15","cited_by_count":3,"area":null},{"id":"2608.23875","title":"AI Finds A Way","created":"2026-08-24","cited_by_count":3,"area":null},{"id":"2509.11398","title":"From Firewalls to Frontiers: AI Red-Teaming is a Domain-Specific Evolution of Cyber Red-Teaming","created":"2025-09-14","cited_by_count":2,"area":null},{"id":"2601.08768","title":"AI as Entertainment","created":"2026-01-13","cited_by_count":2,"area":null},{"id":"2603.08723","title":"Alignment as Iatrogenesis: Pastoral Power, Collective Pathology, and the Structural Limits of Monolingual Safety Evaluation","created":"2026-02-17","cited_by_count":2,"area":null},{"id":"2605.06390","title":"Automated alignment is harder than you think","created":"2026-05-07","cited_by_count":2,"area":null},{"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":"2606.26057","title":"The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems","created":"2026-06-24","cited_by_count":2,"area":null},{"id":"2609.01595","title":"Mechanism Design for Alignment and Control","created":"2026-09-01","cited_by_count":2,"area":null},{"id":"2505.07846","title":"Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models","created":"2025-05-07","cited_by_count":1,"area":null},{"id":"2509.10297","title":"The Morality of Probability: How Implicit Moral Biases in LLMs May Shape the Future of Human-AI Symbiosis","created":"2025-09-12","cited_by_count":1,"area":null},{"id":"2601.10896","title":"DialDefer: A Framework for Detecting and Mitigating LLM Dialogic Deference","created":"2026-01-15","cited_by_count":1,"area":null},{"id":"2602.07259","title":"Incentive-Aware AI Safety via Strategic Resource Allocation: A Stackelberg Security Games Perspective","created":"2026-02-06","cited_by_count":1,"area":null},{"id":"2604.25249","title":"Below-Chance Blindness: Prompted Underperformance in Small LLMs Produces Positional Bias Rather than Answer Avoidance","created":"2026-04-28","cited_by_count":1,"area":null},{"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":"2606.06286","title":"LLMs Can Leak Training Data But Do They Want To? A Propensity-Aware Evaluation of Memorization in LLMs","created":"2026-06-04","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":"2606.26071","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":"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":"2607.13346","title":"The Refusal Residue: When Probes Catch Alignment Faking and When They Don't","created":"2026-07-15","cited_by_count":1,"area":null},{"id":"2608.21766","title":"Evaluation Awareness in Language Models: Representation, Verbalization, and Control","created":"2026-08-22","cited_by_count":1,"area":null},{"id":"2608.27808","title":"CURA: Certified Runtime Alarms for Computer-Use Agents","created":"2026-08-28","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":"2509.05624","title":"Behavioral Inference at Scale: The Fundamental Asymmetry Between Motivations and Belief Systems","created":"2025-09-06","cited_by_count":0,"area":null},{"id":"2510.02840","title":"Take Goodhart Seriously: Principled Limit on General-Purpose AI Optimization","created":"2025-10-03","cited_by_count":0,"area":null},{"id":"2511.01363","title":"Automatic Minds: Cognitive Parallels Between Hypnotic States and Large Language Model Processing","created":"2025-11-03","cited_by_count":0,"area":null},{"id":"2511.21838","title":"Dark Speculation: Combining Qualitative and Quantitative Understanding in Frontier AI Risk Analysis","created":"2025-11-26","cited_by_count":0,"area":null},{"id":"2511.22455","title":"Beyond Real versus Fake Towards Intent-Aware Video Analysis","created":"2025-11-27","cited_by_count":0,"area":null},{"id":"2512.03001","title":"Invasive Context Engineering to Control Large Language Models","created":"2025-12-02","cited_by_count":0,"area":null},{"id":"2601.17577","title":"Status Hierarchies in Language Models","created":"2026-01-24","cited_by_count":0,"area":null},{"id":"2602.14529","title":"Disentangling Deception and Hallucination Failures in LLMs","created":"2026-02-16","cited_by_count":0,"area":null},{"id":"2603.13378","title":"Do Large Language Models Get Caught in Hofstadter-Mobius Loops?","created":"2026-03-10","cited_by_count":0,"area":null},{"id":"2604.00324","title":"The Persistent Vulnerability of Aligned AI Systems","created":"2026-03-31","cited_by_count":0,"area":null},{"id":"2604.05826","title":"Reciprocal Trust and Distrust in Artificial Intelligence Systems: The Hard Problem of Regulation","created":"2026-04-07","cited_by_count":0,"area":null},{"id":"2604.13301","title":"Honeypot Protocol","created":"2026-04-14","cited_by_count":0,"area":null},{"id":"2604.26206","title":"Option-Order Randomisation Reveals a Distributional Position Attractor in Prompted Sandbagging","created":"2026-04-29","cited_by_count":0,"area":null},{"id":"2604.27249","title":"Instruction Complexity Induces Positional Collapse in Adversarial LLM Evaluation","created":"2026-04-29","cited_by_count":0,"area":null},{"id":"2605.02398","title":"The Compliance Trap: How Structural Constraints Degrade Frontier AI Metacognition Under Adversarial Pressure","created":"2026-05-04","cited_by_count":0,"area":null},{"id":"2605.16872","title":"Some[Body] Must Receive That Pain for Agent Accountability","created":"2026-05-16","cited_by_count":0,"area":null},{"id":"2605.19270","title":"DECOR: Auditing LLM Deception via Information Manipulation Theory","created":"2026-05-19","cited_by_count":0,"area":null},{"id":"2605.27681","title":"Behavioural Analysis of Alignment Faking","created":"2026-05-26","cited_by_count":0,"area":null},{"id":"2605.30406","title":"AI Loss of Control Incident Management: Response & Resilience","created":"2026-05-28","cited_by_count":0,"area":null},{"id":"2606.08629","title":"Sycophancy Towards Researchers Drives Performative Misalignment","created":"2026-06-07","cited_by_count":0,"area":null},{"id":"2606.10456","title":"The Distributed Detectability Band Against Marginal-Preserving Attacks","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2606.12442","title":"Reframing AI Loss of Control: What Control Is, How to Have It, How to Lose It","created":"2026-05-19","cited_by_count":0,"area":null},{"id":"2606.16914","title":"Greed Is Learned: Visible Incentives as Reward-Hacking Triggers","created":"2026-06-15","cited_by_count":0,"area":null},{"id":"2606.20610","title":"Signals in the Noise: Open Source Intelligence (OSINT) for AI Loss of Control Detection","created":"2026-05-22","cited_by_count":0,"area":null},{"id":"2606.20663","title":"DrugBench: Evaluating AI Control Protocols for Medication Harm Mitigation","created":"2026-06-10","cited_by_count":0,"area":null},{"id":"2606.28863","title":"Defeat Devices in AI Systems","created":"2026-06-27","cited_by_count":0,"area":null},{"id":"2607.24758","title":"Do Models Fake Alignment Without Clear Consequences?","created":"2026-05-28","cited_by_count":0,"area":null},{"id":"2607.24769","title":"LLM Scheming Inversely Scales with Pretraining Language Coverage","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2608.11008","title":"Templated or fully synthetic? Prompt construction as a confound in measuring LLM political stance beyond writing assistance","created":"2026-08-11","cited_by_count":0,"area":null}],"next_cursor":"100"}