{"total":91,"rows":[{"id":"2403.03218","title":"The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning","created":"2024-03-05","cited_by_count":509,"area":"safety"},{"id":"2406.04313","title":"Improving Alignment and Robustness with Circuit Breakers","created":"2024-06-06","cited_by_count":322,"area":"safety"},{"id":"2407.15549","title":"Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs","created":"2024-07-22","cited_by_count":173,"area":"safety"},{"id":"2402.16835","title":"Eight Methods to Evaluate Robust Unlearning in LLMs","created":"2024-02-26","cited_by_count":163,"area":"safety"},{"id":"2510.09023","title":"The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections","created":"2025-10-10","cited_by_count":129,"area":"safety"},{"id":"2409.18025","title":"An Adversarial Perspective on Machine Unlearning for AI Safety","created":"2024-09-26","cited_by_count":123,"area":"safety"},{"id":"2502.17420","title":"The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence","created":"2025-02-24","cited_by_count":115,"area":"alignment"},{"id":"2405.15589","title":"Efficient Adversarial Training in LLMs with Continuous Attacks","created":"2024-05-24","cited_by_count":113,"area":"safety"},{"id":"2404.15146","title":"Rethinking LLM Memorization through the Lens of Adversarial Compression","created":"2024-04-23","cited_by_count":94,"area":"lr"},{"id":"2403.05030","title":"Defending Against Unforeseen Failure Modes with Latent Adversarial Training","created":"2024-03-08","cited_by_count":88,"area":"safety"},{"id":"2404.12038","title":"Uncovering Safety Risks of Large Language Models through Concept Activation Vector","created":"2024-04-18","cited_by_count":67,"area":"alignment"},{"id":"2403.13682","title":"Threats, Attacks, and Defenses in Machine Unlearning: A Survey","created":"2024-03-20","cited_by_count":39,"area":"safety"},{"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":"2407.16997","title":"Revisiting Who's Harry Potter: Towards Targeted Unlearning from a Causal Intervention Perspective","created":"2024-07-24","cited_by_count":36,"area":"safety"},{"id":"2504.01550","title":"Representation Bending for Large Language Model Safety","created":"2025-04-02","cited_by_count":30,"area":"safety"},{"id":"2407.07342","title":"Multilingual Blending: LLM Safety Alignment Evaluation with Language Mixture","created":"2024-07-10","cited_by_count":22,"area":"alignment"},{"id":"2410.03523","title":"A Probabilistic Perspective on Unlearning and Alignment for Large Language Models","created":"2024-10-04","cited_by_count":21,"area":"alignment"},{"id":"2411.02631","title":"Extracting Unlearned Information from LLMs with Activation Steering","created":"2024-11-04","cited_by_count":21,"area":"safety"},{"id":"2603.06594","title":"A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness","created":"2026-02-04","cited_by_count":20,"area":"safety"},{"id":"2410.16531","title":"Bayesian scaling laws for in-context learning","created":"2024-10-21","cited_by_count":19,"area":"safety"},{"id":"2503.02574","title":"LLM-Safety Evaluations Lack Robustness","created":"2025-03-04","cited_by_count":18,"area":"alignment"},{"id":"2506.07171","title":"RULE: Reinforcement UnLEarning Achieves Forget-Retain Pareto Optimality","created":"2025-06-08","cited_by_count":16,"area":"safety"},{"id":"2502.17254","title":"REINFORCE Adversarial Attacks on Large Language Models: An Adaptive, Distributional, and Semantic Objective","created":"2025-02-24","cited_by_count":15,"area":"alignment"},{"id":"2506.15699","title":"BLUR: A Benchmark for LLM Unlearning Robust to Forget-Retain Overlap","created":"2025-05-28","cited_by_count":15,"area":"safety"},{"id":"2502.04204","title":"Short-length Adversarial Training Helps LLMs Defend Long-length Jailbreak Attacks: Theoretical and Empirical Evidence","created":"2025-02-06","cited_by_count":14,"area":"safety"},{"id":"2602.22983","title":"Obscure but Effective: Classical Chinese Jailbreak Prompt Optimization via Bio-Inspired Search","created":"2026-02-26","cited_by_count":13,"area":"safety"},{"id":"2410.19160","title":"Adversarial Attacks on Large Language Models Using Regularized Relaxation","created":"2024-10-24","cited_by_count":12,"area":null},{"id":"2502.16184","title":"Robustness and Cybersecurity in the EU Artificial Intelligence Act","created":"2025-02-22","cited_by_count":11,"area":null},{"id":"2505.17406","title":"Robust Answers, Fragile Logic: Probing the Decoupling Hypothesis in LLM Reasoning","created":"2025-05-23","cited_by_count":11,"area":"lr"},{"id":"2506.09227","title":"SoK: Machine Unlearning for Large Language Models","created":"2025-06-10","cited_by_count":11,"area":"safety"},{"id":"2502.00406","title":"Agents Are All You Need for LLM Unlearning","created":"2025-02-01","cited_by_count":10,"area":"agents"},{"id":"2502.09755","title":"Jailbreak Attack Initializations as Extractors of Compliance Directions","created":"2025-02-13","cited_by_count":10,"area":"safety"},{"id":"2507.02956","title":"A Representation Engineering Perspective on the Effectiveness of Multi-Turn Jailbreaks","created":"2025-06-29","cited_by_count":10,"area":"safety"},{"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":"2506.11938","title":"Improving Large Language Model Safety with Contrastive Representation Learning","created":"2025-06-13","cited_by_count":8,"area":"mechinterp"},{"id":"2407.08970","title":"Self-interpreting Adversarial Images","created":"2024-07-12","cited_by_count":7,"area":"mllm"},{"id":"2505.17160","title":"Harry Potter is Still Here! Probing Knowledge Leakage in Targeted Unlearned Large Language Models via Automated Adversarial Prompting","created":"2025-05-22","cited_by_count":7,"area":"safety"},{"id":"2505.23270","title":"Does Machine Unlearning Truly Remove Knowledge?","created":"2025-05-29","cited_by_count":7,"area":"safety"},{"id":"2507.04219","title":"Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs","created":"2025-07-06","cited_by_count":7,"area":"synth"},{"id":"2509.01631","title":"Unraveling LLM Jailbreaks Through Safety Knowledge Neurons","created":"2025-09-01","cited_by_count":7,"area":"safety"},{"id":"2511.00203","title":"Diffusion LLMs are Natural Adversaries for any LLM","created":"2025-10-31","cited_by_count":7,"area":"safety"},{"id":"2412.08099","title":"Adversarial Vulnerabilities in Large Language Models for Time Series Forecasting","created":"2024-12-11","cited_by_count":6,"area":null},{"id":"2511.04316","title":"AdversariaLLM: A Unified and Modular Toolbox for LLM Robustness Research","created":"2025-11-06","cited_by_count":6,"area":"safety"},{"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":"2510.04891","title":"SocialHarmBench: Revealing LLM Vulnerabilities to Socially Harmful Requests","created":"2025-10-06","cited_by_count":5,"area":"safety"},{"id":"2410.04234","title":"Functional Homotopy: Smoothing Discrete Optimization via Continuous Parameters for LLM Jailbreak Attacks","created":"2024-10-05","cited_by_count":4,"area":"safety"},{"id":"2503.20257","title":"How Secure is Forgetting? Linking Machine Unlearning to Machine Learning Attacks","created":"2025-03-26","cited_by_count":4,"area":"safety"},{"id":"2505.09820","title":"Adversarial Attack on Large Language Models using Exponentiated Gradient Descent","created":"2025-05-14","cited_by_count":4,"area":"safety"},{"id":"2506.17265","title":"SUA: Stealthy Multimodal Large Language Model Unlearning Attack","created":"2025-06-10","cited_by_count":4,"area":"mllm"},{"id":"2510.25117","title":"A Survey on Unlearning in Large Language Models","created":"2025-10-29","cited_by_count":4,"area":"safety"},{"id":"2602.06248","title":"REBEL: Hidden Knowledge Recovery via Evolutionary-Based Evaluation Loop","created":"2026-02-05","cited_by_count":4,"area":"safety"},{"id":"2502.10487","title":"Fast Proxies for LLM Robustness Evaluation","created":"2025-02-14","cited_by_count":3,"area":"safety"},{"id":"2506.18325","title":"NSFW-Classifier Guided Prompt Sanitization for Safe Text-to-Image Generation","created":"2025-06-23","cited_by_count":3,"area":"safety"},{"id":"2406.13283","title":"Large-Scale Dataset Pruning in Adversarial Training through Data Importance Extrapolation","created":"2024-06-19","cited_by_count":2,"area":null},{"id":"2407.13796","title":"Continuous Embedding Attacks via Clipped Inputs in Jailbreaking Large Language Models","created":"2024-07-16","cited_by_count":2,"area":"safety"},{"id":"2502.15836","title":"Soft Token Attacks Cannot Reliably Audit Unlearning in Large Language Models","created":"2025-02-20","cited_by_count":2,"area":"safety"},{"id":"2502.16708","title":"Exploring Incremental Unlearning: Techniques, Challenges, and Future Directions","created":"2025-02-23","cited_by_count":2,"area":"safety"},{"id":"2506.10946","title":"GUARD: Guided Unlearning and Retention via Data Attribution for Large Language Models","created":"2025-06-12","cited_by_count":2,"area":"safety"},{"id":"2506.24048","title":"Consensus-based optimization for closed-box adversarial attacks and a connection to evolution strategies","created":"2025-06-30","cited_by_count":2,"area":null},{"id":"2507.08020","title":"Circumventing Safety Alignment in Large Language Models Through Embedding Space Toxicity Attenuation","created":"2025-07-08","cited_by_count":2,"area":"alignment"},{"id":"2508.08629","title":"Securing Educational LLMs: A Generalised Taxonomy of Attacks on LLMs and DREAD Risk Assessment","created":"2025-08-12","cited_by_count":2,"area":"safety"},{"id":"2508.14853","title":"Universal and Transferable Adversarial Attack on Large Language Models Using Exponentiated Gradient Descent","created":"2025-08-20","cited_by_count":2,"area":"alignment"},{"id":"2510.07626","title":"LLM Unlearning Under the Microscope: A Full-Stack View on Methods and Metrics","created":"2025-10-08","cited_by_count":2,"area":"safety"},{"id":"2606.08044","title":"When Behavioral Safety Evaluation Fails: A Representation-Level Perspective","created":"2026-06-06","cited_by_count":2,"area":"safety"},{"id":"2405.20806","title":"The AI Alignment Paradox","created":"2024-05-31","cited_by_count":1,"area":null},{"id":"2408.02651","title":"Can Reinforcement Learning Unlock the Hidden Dangers in Aligned Large Language Models?","created":"2024-08-05","cited_by_count":1,"area":"alignment"},{"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":"2508.10404","title":"Layer-Wise Perturbations via Sparse Autoencoders for Adversarial Text Generation","created":"2025-08-14","cited_by_count":1,"area":"safety"},{"id":"2509.20230","title":"Beyond Sharp Minima: Robust LLM Unlearning via Feedback-Guided Multi-Point Optimization","created":"2025-09-24","cited_by_count":1,"area":"safety"},{"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":"2602.23391","title":"Detoxifying LLMs via Representation Erasure-Based Preference Optimization","created":"2026-02-24","cited_by_count":1,"area":"alignment"},{"id":"2603.04355","title":"Efficient Refusal Ablation in LLM through Optimal Transport","created":"2026-03-04","cited_by_count":1,"area":"alignment"},{"id":"2603.05772","title":"Depth Charge: Jailbreak Large Language Models from Deep Safety Attention Heads","created":"2026-03-06","cited_by_count":1,"area":"safety"},{"id":"2603.13847","title":"Sirens' Whisper: Inaudible Near-Ultrasonic Jailbreaks of Speech-Driven LLMs","created":"2026-03-14","cited_by_count":1,"area":"safety"},{"id":"2604.17053","title":"Jailbreaking Large Language Models with Morality Attacks","created":"2026-04-18","cited_by_count":1,"area":"safety"},{"id":"2509.06338","title":"Embedding Poisoning: Bypassing Safety Alignment via Embedding Semantic Shift","created":"2025-09-08","cited_by_count":0,"area":"alignment"},{"id":"2510.01288","title":"Microsaccade-Inspired Probing: Positional Encoding Perturbations Reveal LLM Misbehaviours","created":"2025-10-01","cited_by_count":0,"area":null},{"id":"2511.11667","title":"Beyond Superficial Forgetting: Thorough Unlearning through Knowledge Density Estimation and Block Re-insertion","created":"2025-11-11","cited_by_count":0,"area":"safety"},{"id":"2602.15238","title":"Closing the Distribution Gap in Adversarial Training for LLMs","created":"2026-02-16","cited_by_count":0,"area":null},{"id":"2604.10403","title":"Latent Instruction Representation Alignment: defending against jailbreaks, backdoors and undesired knowledge in LLMs","created":"2026-04-12","cited_by_count":0,"area":"safety"},{"id":"2604.24983","title":"Adaptive Prompt Embedding Optimization for LLM Jailbreaking","created":"2026-04-27","cited_by_count":0,"area":"safety"},{"id":"2605.24552","title":"Ellipsoid Control: A White-list Jailbreak Defense via Benign Latent Modeling","created":"2026-05-23","cited_by_count":0,"area":"safety"},{"id":"2606.12978","title":"Trajectory-Level Redirection Attacks on Vision-Language-Action Models","created":"2026-06-11","cited_by_count":0,"area":"vla"},{"id":"2606.22779","title":"DE-FIVE: Detecting Malicious Image Prompts via Fourier Features and Image Vector Embeddings","created":"2026-06-22","cited_by_count":0,"area":"mllm"},{"id":"2607.16227","title":"LLM Unlearning for Cyber Defense: A Survey on Methods, Challenges, and Emerging Threats","created":"2026-06-23","cited_by_count":0,"area":"safety"},{"id":"2607.18228","title":"Logical Judgments Under Pressure: Diagnosing Syllogistic Stability with Learned Soft Prefixes","created":"2026-07-20","cited_by_count":0,"area":"moe"},{"id":"2608.25429","title":"Distance Is Not Enough: Forget-Retain Alignment Gap Predicts LLM Relearning Robustness","created":"2026-08-26","cited_by_count":0,"area":"safety"},{"id":"2609.16229","title":"Test-Time Unlearning via Sparse Autoencoder","created":"2026-09-14","cited_by_count":0,"area":"safety"},{"id":"2609.32116","title":"Escaping Alignment: A Physical Trap Model of Best-of-N Jailbreaking","created":"2026-09-26","cited_by_count":0,"area":"safety"},{"id":"2610.02418","title":"Mitigating Private Data Leakage in LLMs with Whiteout","created":"2026-10-01","cited_by_count":0,"area":"safety"},{"id":"2610.18127","title":"Monitoring latent reasoning in looped language models","created":"2026-10-13","cited_by_count":0,"area":"lr"}],"next_cursor":null}