{"total":76,"rows":[{"id":"2409.20089","title":"Robust LLM safeguarding via refusal feature adversarial training","created":"2024-09-30","cited_by_count":64,"area":null},{"id":"2404.12241","title":"Introducing v0.5 of the AI Safety Benchmark from MLCommons","created":"2024-04-18","cited_by_count":62,"area":null},{"id":"2412.07097","title":"On Evaluating the Durability of Safeguards for Open-Weight LLMs","created":"2024-12-10","cited_by_count":52,"area":null},{"id":"2508.09224","title":"From Hard Refusals to Safe-Completions: Toward Output-Centric Safety Training","created":"2025-08-12","cited_by_count":52,"area":null},{"id":"2410.07137","title":"Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates","created":"2024-10-09","cited_by_count":45,"area":null},{"id":"2502.05209","title":"Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities","created":"2025-02-03","cited_by_count":39,"area":null},{"id":"2407.18369","title":"AI Safety in Generative AI Large Language Models: A Survey","created":"2024-07-06","cited_by_count":33,"area":null},{"id":"2502.12964","title":"Trust Me, I'm Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answer","created":"2025-02-18","cited_by_count":27,"area":null},{"id":"2506.11094","title":"The Scales of Justitia: A Comprehensive Survey on Safety Evaluation of LLMs","created":"2025-06-06","cited_by_count":26,"area":null},{"id":"2412.04653","title":"Hidden in the Noise: Two-Stage Robust Watermarking for Images","created":"2024-12-05","cited_by_count":21,"area":null},{"id":"2406.07954","title":"Dataset and Lessons Learned from the 2024 SaTML LLM Capture-the-Flag Competition","created":"2024-06-12","cited_by_count":20,"area":null},{"id":"2506.00085","title":"COSMIC: Generalized Refusal Direction Identification in LLM Activations","created":"2025-05-30","cited_by_count":19,"area":null},{"id":"2412.06370","title":"Exploring Memorization and Copyright Violation in Frontier LLMs: A Study of the New York Times v. OpenAI 2023 Lawsuit","created":"2024-12-09","cited_by_count":18,"area":null},{"id":"2502.02180","title":"The Elicitation Game: Evaluating Capability Elicitation Techniques","created":"2025-02-04","cited_by_count":18,"area":null},{"id":"2502.02260","title":"Position: Adversarial ML for LLMs Is Not Making Any Progress","created":"2025-02-04","cited_by_count":15,"area":null},{"id":"2408.02509","title":"Black-Box Adversarial Attacks on LLM-Based Code Completion","created":"2024-08-05","cited_by_count":14,"area":null},{"id":"2505.17646","title":"Unveiling the Basin-Like Loss Landscape in Large Language Models","created":"2025-05-23","cited_by_count":13,"area":null},{"id":"2601.21571","title":"Shaping capabilities with token-level data filtering","created":"2026-01-29","cited_by_count":13,"area":null},{"id":"2502.16776","title":"AISafetyLab: A Comprehensive Framework for AI Safety Evaluation and Improvement","created":"2025-02-24","cited_by_count":12,"area":null},{"id":"2509.13281","title":"RepIt: Steering Language Models with Concept-Specific Refusal Vectors","created":"2025-09-16","cited_by_count":12,"area":null},{"id":"2512.16856","title":"Distributional AGI Safety","created":"2025-12-18","cited_by_count":12,"area":null},{"id":"2505.17441","title":"Discovering Forbidden Topics in Language Models","created":"2025-05-23","cited_by_count":9,"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":"2502.16792","title":"The Role of Sparsity for Length Generalization in Transformers","created":"2025-02-24","cited_by_count":8,"area":null},{"id":"2506.06414","title":"Benchmarking Misuse Mitigation Against Covert Adversaries","created":"2025-06-06","cited_by_count":8,"area":null},{"id":"2510.15232","title":"FinTrust: A Comprehensive Benchmark of Trustworthiness Evaluation in Finance Domain","created":"2025-10-17","cited_by_count":8,"area":null},{"id":"2604.14548","title":"VoxSafeBench: Not Just What Is Said, but Who, How, and Where","created":"2026-04-16","cited_by_count":8,"area":null},{"id":"2501.19040","title":"Towards the Worst-case Robustness of Large Language Models","created":"2025-01-31","cited_by_count":7,"area":null},{"id":"2405.20404","title":"JoPA:Explaining Large Language Model's Generation via Joint Prompt Attribution","created":"2024-05-30","cited_by_count":6,"area":null},{"id":"2409.03274","title":"Recent Advances in Attack and Defense Approaches of Large Language Models","created":"2024-09-05","cited_by_count":6,"area":null},{"id":"2505.15323","title":"Improving LLM First-Token Predictions in Multiple-Choice Question Answering via Output Prefilling","created":"2025-05-21","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":"2505.24621","title":"Benchmarking Large Language Models for Cryptanalysis and Side-Channel Vulnerabilities","created":"2025-05-30","cited_by_count":5,"area":null},{"id":"2510.10998","title":"ABLEIST: Intersectional Disability Bias in LLM-Generated Hiring Scenarios","created":"2025-10-13","cited_by_count":5,"area":null},{"id":"2511.18852","title":"FanarGuard: A Culturally-Aware Moderation Filter for Arabic Language Models","created":"2025-11-24","cited_by_count":5,"area":null},{"id":"2409.12367","title":"Extracting Memorized Training Data via Decomposition","created":"2024-09-18","cited_by_count":4,"area":null},{"id":"2506.07330","title":"JavelinGuard: Low-Cost Transformer Architectures for LLM Security","created":"2025-06-09","cited_by_count":4,"area":null},{"id":"2411.19140","title":"Examining Multimodal Gender and Content Bias in ChatGPT-4o","created":"2024-11-28","cited_by_count":3,"area":null},{"id":"2503.16534","title":"Gender and content bias in Large Language Models: a case study on Google Gemini 2.0 Flash Experimental","created":"2025-03-18","cited_by_count":3,"area":null},{"id":"2503.17388","title":"AI Companies Should Report Pre- and Post-Mitigation Safety Evaluations","created":"2025-03-17","cited_by_count":3,"area":null},{"id":"2505.13899","title":"Causes and Consequences of Representational Similarity in Machine Learning Models","created":"2025-05-20","cited_by_count":3,"area":null},{"id":"2602.11528","title":"Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs","created":"2026-02-12","cited_by_count":3,"area":null},{"id":"2604.23459","title":"Architecture Matters for Multi-Agent Security","created":"2026-04-25","cited_by_count":3,"area":null},{"id":"2411.15222","title":"Rethinking the Intermediate Features in Adversarial Attacks: Misleading Robotic Models via Adversarial Distillation","created":"2024-11-21","cited_by_count":2,"area":null},{"id":"2412.20087","title":"On the Validity of Traditional Vulnerability Scoring Systems for Adversarial Attacks against LLMs","created":"2024-12-28","cited_by_count":2,"area":null},{"id":"2603.27332","title":"Unsafe by Reciprocity: How Generation-Understanding Coupling Undermines Safety in Unified Multimodal Models","created":"2026-03-28","cited_by_count":2,"area":null},{"id":"2608.28327","title":"Layered LLM Defenses as an Ensemble: Access Tiers, Inference Cost, and the Measured Failure Correlation Between Defense Layers","created":"2026-08-28","cited_by_count":2,"area":null},{"id":"2502.16366","title":"A Generative Approach to LLM Harmfulness Mitigation with Red Flag Tokens","created":"2025-02-22","cited_by_count":1,"area":null},{"id":"2504.20310","title":"A Cryptographic Perspective on Mitigation vs. Detection in Machine Learning","created":"2025-04-28","cited_by_count":1,"area":null},{"id":"2505.02273","title":"Demystifying optimized prompts in language models","created":"2025-05-04","cited_by_count":1,"area":null},{"id":"2508.08855","title":"BiasGym: A Simple and Generalizable Framework for Analyzing and Removing Biases through Injection","created":"2025-08-12","cited_by_count":1,"area":null},{"id":"2509.10682","title":"LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systems","created":"2025-09-12","cited_by_count":1,"area":null},{"id":"2511.05919","title":"Injecting Falsehoods: Adversarial Man-in-the-Middle Attacks Undermining Factual Recall in LLMs","created":"2025-11-08","cited_by_count":1,"area":null},{"id":"2601.11575","title":"Concept Attractors in LLMs and their Applications","created":"2025-12-30","cited_by_count":1,"area":null},{"id":"2601.23094","title":"Safer Policy Compliance with Dynamic Epistemic Fallback","created":"2026-01-30","cited_by_count":1,"area":null},{"id":"2605.27117","title":"Position: AI Safety Requires Effective Controllability","created":"2026-05-26","cited_by_count":1,"area":null},{"id":"2606.05817","title":"Consistency Training Along the Transformer Stack","created":"2026-06-04","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":"2411.19038","title":"DIESEL -- Dynamic Inference-Guidance via Evasion of Semantic Embeddings in LLMs","created":"2024-11-28","cited_by_count":0,"area":null},{"id":"2503.00224","title":"\\`A la recherche du sens perdu: your favourite LLM might have more to say than you can understand","created":"2025-02-28","cited_by_count":0,"area":null},{"id":"2504.04981","title":"TestDG: Test-time Domain Generalization for Continual Test-time Adaptation","created":"2025-04-07","cited_by_count":0,"area":null},{"id":"2506.06933","title":"Rewriting the Budget: A General Framework for Black-Box Attacks Under Cost Asymmetry","created":"2025-06-07","cited_by_count":0,"area":null},{"id":"2510.11834","title":"Don't Walk the Line: Boundary Guidance for Filtered Generation","created":"2025-10-13","cited_by_count":0,"area":null},{"id":"2602.02258","title":"Alignment-Aware Model Adaptation via Feedback-Guided Optimization","created":"2026-02-02","cited_by_count":0,"area":null},{"id":"2602.15238","title":"Closing the Distribution Gap in Adversarial Training for LLMs","created":"2026-02-16","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":"2604.22089","title":"Ethics Testing: Proactive Identification of Generative AI System Harms","created":"2026-04-23","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":"2606.23671","title":"Can LLMs Reliably Self-Report Adversarial Prefills, and How?","created":"2026-06-22","cited_by_count":0,"area":null},{"id":"2606.30219","title":"EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures","created":"2026-06-29","cited_by_count":0,"area":null},{"id":"2606.30989","title":"Wait, am I Being Fair? Characterizing Deductive Stereotyping and Mitigating It with Fair-GCG","created":"2026-06-30","cited_by_count":0,"area":null},{"id":"2607.28959","title":"Efficient LLM Adversarial Training via Low-Rank Defense and Circuit-Guided Surrogates","created":"2026-07-31","cited_by_count":0,"area":null},{"id":"2608.31105","title":"BLOOM-WILT: Logit Tilting for Behaviour Elicitation in Automated LLM Auditing","created":"2026-08-31","cited_by_count":0,"area":null},{"id":"2609.00519","title":"The Safeguard Worked. Is the LLM System Safer?","created":"2026-09-01","cited_by_count":0,"area":null},{"id":"2609.00760","title":"A Unified Mechanistic Analysis of Knowledge- and Safety-Based Refusals","created":"2026-09-01","cited_by_count":0,"area":null},{"id":"2609.10142","title":"Active Adaptation, Not Static Defense: Temporal Dynamics of Preventative Steering in Adversarial Fine-Tuning","created":"2026-09-09","cited_by_count":0,"area":null}],"next_cursor":null}