{"total":60,"rows":[{"id":"2211.00593","title":"Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small","created":"2022-11-01","cited_by_count":1165,"area":"mechinterp"},{"id":"2211.09527","title":"Ignore Previous Prompt: Attack Techniques For Language Models","created":"2022-11-17","cited_by_count":832,"area":null},{"id":"2304.14997","title":"Towards Automated Circuit Discovery for Mechanistic Interpretability","created":"2023-04-28","cited_by_count":724,"area":"mechinterp"},{"id":"2404.14082","title":"Mechanistic Interpretability for AI Safety -- A Review","created":"2024-04-22","cited_by_count":422,"area":"mechinterp"},{"id":"2210.07242","title":"OpenOOD: Benchmarking Generalized Out-of-Distribution Detection","created":"2022-10-13","cited_by_count":268,"area":null},{"id":"2209.00626","title":"The Alignment Problem from a Deep Learning Perspective","created":"2022-08-30","cited_by_count":258,"area":null},{"id":"2307.10169","title":"Challenges and Applications of Large Language Models","created":"2023-07-19","cited_by_count":254,"area":null},{"id":"2306.04488","title":"Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards","created":"2023-06-07","cited_by_count":243,"area":"alignment"},{"id":"2306.12001","title":"An Overview of Catastrophic AI Risks","created":"2023-06-21","cited_by_count":222,"area":null},{"id":"2304.03279","title":"Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmark","created":"2023-04-06","cited_by_count":182,"area":null},{"id":"2401.12187","title":"WARM: On the Benefits of Weight Averaged Reward Models","created":"2024-01-22","cited_by_count":147,"area":"alignment"},{"id":"2210.14891","title":"Broken Neural Scaling Laws","created":"2022-10-26","cited_by_count":109,"area":null},{"id":"2307.03718","title":"Frontier AI Regulation: Managing Emerging Risks to Public Safety","created":"2023-07-06","cited_by_count":98,"area":null},{"id":"2403.08715","title":"SOTOPIA-$\\pi$: Interactive Learning of Socially Intelligent Language Agents","created":"2024-03-13","cited_by_count":93,"area":"agents"},{"id":"2304.10970","title":"Can GPT-4 Perform Neural Architecture Search?","created":"2023-04-21","cited_by_count":74,"area":null},{"id":"2206.15474","title":"Forecasting Future World Events with Neural Networks","created":"2022-06-30","cited_by_count":63,"area":null},{"id":"2407.21792","title":"Safetywashing: Do AI Safety Benchmarks Actually Measure Safety Progress?","created":"2024-07-31","cited_by_count":61,"area":null},{"id":"2306.09983","title":"Evaluating Superhuman Models with Consistency Checks","created":"2023-06-16","cited_by_count":47,"area":null},{"id":"2406.16768","title":"WARP: On the Benefits of Weight Averaged Rewarded Policies","created":"2024-06-24","cited_by_count":45,"area":"alignment"},{"id":"2407.21794","title":"Generalized Out-of-Distribution Detection and Beyond in Vision Language Model Era: A Survey","created":"2024-07-31","cited_by_count":36,"area":"mllm"},{"id":"2505.12923","title":"The Traitors: Deception and Trust in Multi-Agent Language Model Simulations","created":"2025-05-19","cited_by_count":34,"area":"agents"},{"id":"2303.08721","title":"Artificial Influence: An Analysis Of AI-Driven Persuasion","created":"2023-03-15","cited_by_count":33,"area":null},{"id":"2405.10313","title":"How Far Are We From AGI: Are LLMs All We Need?","created":"2024-05-16","cited_by_count":31,"area":null},{"id":"2401.07836","title":"Two Types of AI Existential Risk: Decisive and Accumulative","created":"2024-01-15","cited_by_count":22,"area":null},{"id":"2506.01080","title":"The Coming Crisis of Multi-Agent Misalignment: AI Alignment Must Be a Dynamic and Social Process","created":"2025-06-01","cited_by_count":21,"area":"alignment"},{"id":"2211.06519","title":"The Expertise Problem: Learning from Specialized Feedback","created":"2022-11-12","cited_by_count":20,"area":"alignment"},{"id":"2404.14068","title":"Holistic Safety and Responsibility Evaluations of Advanced AI Models","created":"2024-04-22","cited_by_count":19,"area":null},{"id":"2503.02574","title":"LLM-Safety Evaluations Lack Robustness","created":"2025-03-04","cited_by_count":18,"area":"alignment"},{"id":"2308.10261","title":"How Good Are LLMs at Out-of-Distribution Detection?","created":"2023-08-20","cited_by_count":17,"area":null},{"id":"2502.08301","title":"Compromising Honesty and Harmlessness in Language Models via Deception Attacks","created":"2025-02-12","cited_by_count":17,"area":null},{"id":"2311.04131","title":"Towards Interpretable Sequence Continuation: Analyzing Shared Circuits in Large Language Models","created":"2023-11-07","cited_by_count":15,"area":"mechinterp"},{"id":"2403.20331","title":"Unsolvable Problem Detection: Robust Understanding Evaluation for Large Multimodal Models","created":"2024-03-29","cited_by_count":13,"area":"mllm"},{"id":"2407.18676","title":"Right Now, Wrong Then: Non-Stationary Direct Preference Optimization under Preference Drift","created":"2024-07-26","cited_by_count":13,"area":"alignment"},{"id":"2506.06166","title":"The Lock-in Hypothesis: Stagnation by Algorithm","created":"2025-06-06","cited_by_count":11,"area":null},{"id":"2210.10039","title":"How Would The Viewer Feel? Estimating Wellbeing From Video Scenarios","created":"2022-10-18","cited_by_count":9,"area":null},{"id":"2210.12429","title":"Robots-Dont-Cry: Understanding Falsely Anthropomorphic Utterances in Dialog Systems","created":"2022-10-22","cited_by_count":8,"area":null},{"id":"2405.18979","title":"MANO: Exploiting Matrix Norm for Unsupervised Accuracy Estimation Under Distribution Shifts","created":"2024-05-29","cited_by_count":8,"area":null},{"id":"2411.08243","title":"Beyond the Safety Bundle: Auditing the Helpful and Harmless Dataset","created":"2024-11-12","cited_by_count":8,"area":null},{"id":"2502.11910","title":"Adversarial Alignment for LLMs Requires Simpler, Reproducible, and More Measurable Objectives","created":"2025-02-17","cited_by_count":8,"area":null},{"id":"2312.14751","title":"Hazards from Increasingly Accessible Fine-Tuning of Downloadable Foundation Models","created":"2023-12-22","cited_by_count":7,"area":null},{"id":"2309.08636","title":"ChatGPT v Bard v Bing v Claude 2 v Aria v human-expert. How good are AI chatbots at scientific writing?","created":"2023-09-14","cited_by_count":6,"area":null},{"id":"2311.17228","title":"Survey on AI Ethics: A Socio-technical Perspective","created":"2023-11-28","cited_by_count":4,"area":null},{"id":"2505.21800","title":"From Directions to Cones: Exploring Multidimensional Representations of Propositional Facts in LLMs","created":"2025-05-27","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":"2304.09826","title":"Fairness in AI and Its Long-Term Implications on Society","created":"2023-04-16","cited_by_count":3,"area":null},{"id":"2308.14969","title":"Uncovering the Hidden Cost of Model Compression","created":"2023-08-29","cited_by_count":3,"area":null},{"id":"2501.15054","title":"Unraveling Token Prediction Refinement and Identifying Essential Layers in Language Models","created":"2025-01-25","cited_by_count":3,"area":null},{"id":"2504.07104","title":"Relevance Isn't All You Need: Scaling RAG Systems With Inference-Time Compute Via Multi-Criteria Reranking","created":"2025-03-14","cited_by_count":3,"area":"lr"},{"id":"2407.03489","title":"FlowCon: Out-of-Distribution Detection using Flow-Based Contrastive Learning","created":"2024-07-03","cited_by_count":2,"area":null},{"id":"2410.07456","title":"SAGE: Scalable Ground Truth Evaluations for Large Sparse Autoencoders","created":"2024-10-09","cited_by_count":2,"area":"mechinterp"},{"id":"2504.14496","title":"Functional Abstraction of Knowledge Recall in Large Language Models","created":"2025-04-20","cited_by_count":1,"area":null},{"id":"2504.14838","title":"Establishing Reliability Metrics for Reward Models in Large Language Models","created":"2025-04-21","cited_by_count":1,"area":"alignment"},{"id":"2601.10802","title":"ICONIC-444: A 3.1-Million-Image Dataset for OOD Detection Research","created":"2026-01-15","cited_by_count":1,"area":null},{"id":"2408.06736","title":"Speculations on Uncertainty and Humane Algorithms","created":"2024-08-13","cited_by_count":0,"area":null},{"id":"2410.10984","title":"Data-Aware Training Quality Monitoring and Certification for Reliable Deep Learning","created":"2024-10-14","cited_by_count":0,"area":null},{"id":"2412.03058","title":"Revisiting Energy-Based Model for Out-of-Distribution Detection","created":"2024-12-04","cited_by_count":0,"area":null},{"id":"2507.08014","title":"Mass-Scale Analysis of In-the-Wild Conversations Reveals Complexity Bounds on LLM Jailbreaking","created":"2025-07-06","cited_by_count":0,"area":"safety"},{"id":"2508.19500","title":"Servant, Stalker, Predator: How An Honest, Helpful, And Harmless (3H) Agent Unlocks Adversarial Skills","created":"2025-08-27","cited_by_count":0,"area":"agents"},{"id":"2602.01644","title":"From Perception to Action: Spatial AI Agents and World Models","created":"2026-02-02","cited_by_count":0,"area":"wm"},{"id":"2609.06785","title":"Unified Multi-Layer Subspace Modeling for Cross-Domain OOD Detection","created":"2026-09-06","cited_by_count":0,"area":null}],"next_cursor":null}