{"total":61,"rows":[{"id":"2503.03750","title":"The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems","created":"2025-03-05","cited_by_count":72,"area":"agents"},{"id":"2501.18841","title":"Trading Inference-Time Compute for Adversarial Robustness","created":"2025-01-31","cited_by_count":61,"area":"lr"},{"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":"2412.17686","title":"Large Language Model Safety: A Holistic Survey","created":"2024-12-23","cited_by_count":59,"area":"safety"},{"id":"2602.17753","title":"The 2025 AI Agent Index: Documenting Technical and Safety Features of Deployed Agentic AI Systems","created":"2026-02-19","cited_by_count":30,"area":"agents"},{"id":"2501.07238","title":"Lessons From Red Teaming 100 Generative AI Products","created":"2025-01-13","cited_by_count":23,"area":"safety"},{"id":"2410.01294","title":"Endless Jailbreaks with Bijection Learning","created":"2024-10-02","cited_by_count":19,"area":"safety"},{"id":"2502.15799","title":"Investigating the Impact of Quantization Methods on the Safety and Reliability of Large Language Models","created":"2025-02-18","cited_by_count":18,"area":"lr"},{"id":"2506.20702","title":"The Singapore Consensus on Global AI Safety Research Priorities","created":"2025-06-25","cited_by_count":10,"area":null},{"id":"2505.11926","title":"SafeVid: Toward Safety Aligned Video Large Multimodal Models","created":"2025-05-17","cited_by_count":9,"area":"mllm"},{"id":"2509.00591","title":"Probe-Rewrite-Evaluate: A Workflow for Reliable Benchmarks and Quantifying Evaluation Awareness","created":"2025-08-30","cited_by_count":9,"area":null},{"id":"2506.06414","title":"Benchmarking Misuse Mitigation Against Covert Adversaries","created":"2025-06-06","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":"2604.16403","title":"Computational Hermeneutics: Evaluating generative AI as a cultural technology","created":"2026-03-31","cited_by_count":8,"area":null},{"id":"2412.15035","title":"LLMs Lost in Translation: M-ALERT uncovers Cross-Linguistic Safety Inconsistencies","created":"2024-12-19","cited_by_count":7,"area":null},{"id":"2503.05628","title":"Superintelligence Strategy: Expert Version","created":"2025-03-07","cited_by_count":7,"area":null},{"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":"2508.05464","title":"Bench-2-CoP: Can We Trust Benchmarking for EU AI Compliance?","created":"2025-08-07","cited_by_count":6,"area":null},{"id":"2411.11853","title":"Chat Bankman-Fried: an Exploration of LLM Alignment in Finance","created":"2024-11-01","cited_by_count":5,"area":"alignment"},{"id":"2506.15751","title":"Sysformer: Safeguarding Frozen Large Language Models with Adaptive System Prompts","created":"2025-06-18","cited_by_count":5,"area":"safety"},{"id":"2601.10520","title":"Breaking Up with Normatively Monolithic Agency with GRACE: A Reason-Based Neuro-Symbolic Architecture for Safe and Ethical AI Alignment","created":"2026-01-15","cited_by_count":5,"area":"lr"},{"id":"2506.10378","title":"Discovering Hierarchical Latent Capabilities of Language Models via Causal Representation Learning","created":"2025-06-12","cited_by_count":4,"area":null},{"id":"2507.06277","title":"The Prompt War: How AI Decides on a Military Intervention","created":"2025-07-08","cited_by_count":4,"area":null},{"id":"2509.16332","title":"Psychometric Personality Shaping Modulates Capabilities and Safety in Language Models","created":"2025-09-19","cited_by_count":4,"area":null},{"id":"2605.16282","title":"Taxonomy and Consistency Analysis of Safety Benchmarks for AI Agents","created":"2026-04-11","cited_by_count":4,"area":"agents"},{"id":"2606.09809","title":"Evaluation Cards: An Interpretive Layer for AI Evaluation Reporting","created":"2026-06-08","cited_by_count":4,"area":null},{"id":"2606.24014","title":"Reinforcement Learning Towards Broadly and Persistently Beneficial Models","created":"2026-06-22","cited_by_count":4,"area":"alignment"},{"id":"2502.15873","title":"Practical Principles for AI Cost and Compute Accounting","created":"2025-02-21","cited_by_count":3,"area":null},{"id":"2502.18339","title":"Correlating and Predicting Human Evaluations of Language Models from Natural Language Processing Benchmarks","created":"2025-02-24","cited_by_count":3,"area":null},{"id":"2505.17760","title":"But what is your honest answer? Aiding LLM-judges with honest alternatives using steering vectors","created":"2025-05-23","cited_by_count":3,"area":"mechinterp"},{"id":"2603.26676","title":"Evaluating Human-AI Safety: A Framework for Measuring Harmful Capability Uplift","created":"2026-03-06","cited_by_count":3,"area":null},{"id":"2605.04135","title":"Frontier Lag: A Bibliometric Audit of Capability Misrepresentation in Academic AI Evaluation","created":"2026-05-05","cited_by_count":3,"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":"2504.14039","title":"MEQA: A Meta-Evaluation Framework for Question & Answer LLM Benchmarks","created":"2025-04-18","cited_by_count":2,"area":null},{"id":"2505.14215","title":"Safety Degradation in AI Agents","created":"2025-05-20","cited_by_count":2,"area":"agents"},{"id":"2510.13852","title":"ConsistencyAI: A Benchmark to Assess LLMs' Factual Consistency When Responding to Different Demographic Groups","created":"2025-10-11","cited_by_count":2,"area":null},{"id":"2512.10058","title":"Mind the Gap! Pathways Towards Unifying AI Safety and Ethics Research","created":"2025-12-10","cited_by_count":2,"area":null},{"id":"2512.10687","title":"Safe for Whom? Rethinking How We Evaluate the Safety of LLMs for Real Users","created":"2025-12-11","cited_by_count":2,"area":"lr"},{"id":"2604.06233","title":"Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules","created":"2026-04-03","cited_by_count":2,"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":"2508.06111","title":"SKATE, a Scalable Tournament Eval: Weaker LLMs differentiate between stronger ones using verifiable challenges","created":"2025-08-08","cited_by_count":1,"area":null},{"id":"2602.14844","title":"Interactionless Inverse Reinforcement Learning: A Data-Centric Framework for Durable Alignment","created":"2026-02-16","cited_by_count":1,"area":"offlinerl"},{"id":"2602.20813","title":"Pressure Reveals Character: Behavioural Alignment Evaluation at Depth","created":"2026-02-24","cited_by_count":1,"area":null},{"id":"2604.09189","title":"Do LLMs Follow Their Own Rules? A Reflexive Audit of Self-Stated Safety Policies","created":"2026-04-10","cited_by_count":1,"area":"alignment"},{"id":"2606.10154","title":"Quality Is Not a Safety Proxy Under Quantization","created":"2026-06-08","cited_by_count":1,"area":null},{"id":"2608.17183","title":"Benchmarking the Benchmarks: Evaluating Automated Safety Benchmarks for Small Language Models","created":"2026-08-17","cited_by_count":1,"area":null},{"id":"2507.06282","title":"The bitter lesson of misuse detection","created":"2025-07-08","cited_by_count":0,"area":"safety"},{"id":"2507.06434","title":"Deprecating Benchmarks: Criteria and Framework","created":"2025-07-08","cited_by_count":0,"area":null},{"id":"2510.06790","title":"Get RICH or Die Scaling: Profitably Trading Inference Compute for Robustness","created":"2025-10-08","cited_by_count":0,"area":"lr"},{"id":"2602.18456","title":"Beyond single-channel agentic benchmarking","created":"2026-02-05","cited_by_count":0,"area":"agents"},{"id":"2603.05175","title":"Incentive Aware AI Regulations: A Credal Characterisation","created":"2026-03-05","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":"2604.12500","title":"Safety Training Modulates Harmful Misalignment Under On-Policy RL, But Direction Depends on Environment Design","created":"2026-04-14","cited_by_count":0,"area":null},{"id":"2606.00003","title":"Learning from Mistakes: Can LLM Self-Recover after Misalignment?","created":"2026-03-25","cited_by_count":0,"area":"safety"},{"id":"2607.05407","title":"Position: Preventing AI-Generated CSAM Necessitates New Approaches to AI Safety","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2607.09306","title":"Exposure is not manifestation: measurement target and output resolution jointly determine which behavioural-faithfulness evaluator wins","created":"2026-07-10","cited_by_count":0,"area":null},{"id":"2608.18086","title":"Position: Current Model Cards Are Insufficient for Downstream Governance of Open-Weight Foundation Models","created":"2026-06-05","cited_by_count":0,"area":null},{"id":"2608.27009","title":"The Guard That Cried Wolf: How Scary Words Make Agent Guardrails Refuse Legitimate Actions","created":"2026-08-27","cited_by_count":0,"area":"agents"},{"id":"2610.01058","title":"MOMAT: Mixture of Multiple Atlases for Low-Power Jailbreak Defense of Quantized LLMs","created":"2026-10-01","cited_by_count":0,"area":"safety"},{"id":"2610.08540","title":"Toward Alignment Scaling Laws: A Framework and First Preregistered Measurements","created":"2026-10-06","cited_by_count":0,"area":"safety"},{"id":"2610.10232","title":"LLM Persuasion Is in the Eye of the Evaluation","created":"2026-10-07","cited_by_count":0,"area":null}],"next_cursor":null}