{"total":74,"rows":[{"id":"2406.11717","title":"Refusal in Language Models Is Mediated by a Single Direction","created":"2024-06-17","cited_by_count":1032,"area":"safety"},{"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":"2407.04295","title":"Jailbreak Attacks and Defenses Against Large Language Models: A Survey","created":"2024-07-05","cited_by_count":230,"area":"safety"},{"id":"2310.10844","title":"Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks","created":"2023-10-16","cited_by_count":194,"area":"safety"},{"id":"2402.08679","title":"COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability","created":"2024-02-13","cited_by_count":175,"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":"2408.00761","title":"Tamper-Resistant Safeguards for Open-Weight LLMs","created":"2024-08-01","cited_by_count":138,"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":"2404.01099","title":"What is in Your Safe Data? Identifying Benign Data that Breaks Safety","created":"2024-04-01","cited_by_count":118,"area":"safety"},{"id":"2401.17256","title":"Weak-to-Strong Jailbreaking on Large Language Models","created":"2024-01-30","cited_by_count":107,"area":"safety"},{"id":"2405.14577","title":"Representation Noising: A Defence Mechanism Against Harmful Finetuning","created":"2024-05-23","cited_by_count":96,"area":"safety"},{"id":"2402.14020","title":"Coercing LLMs to do and reveal (almost) anything","created":"2024-02-21","cited_by_count":89,"area":"safety"},{"id":"2403.05030","title":"Defending Against Unforeseen Failure Modes with Latent Adversarial Training","created":"2024-03-08","cited_by_count":88,"area":"safety"},{"id":"2402.16717","title":"CodeChameleon: Personalized Encryption Framework for Jailbreaking Large Language Models","created":"2024-02-26","cited_by_count":76,"area":"safety"},{"id":"2406.01288","title":"Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses","created":"2024-06-03","cited_by_count":70,"area":"safety"},{"id":"2403.12171","title":"EasyJailbreak: A Unified Framework for Jailbreaking Large Language Models","created":"2024-03-18","cited_by_count":60,"area":"safety"},{"id":"2412.17686","title":"Large Language Model Safety: A Holistic Survey","created":"2024-12-23","cited_by_count":59,"area":"safety"},{"id":"2406.17092","title":"BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models","created":"2024-06-24","cited_by_count":58,"area":"safety"},{"id":"2403.04893","title":"A Safe Harbor for AI Evaluation and Red Teaming","created":"2024-03-07","cited_by_count":47,"area":"safety"},{"id":"2312.00027","title":"Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections","created":"2023-11-15","cited_by_count":39,"area":"safety"},{"id":"2410.15236","title":"Jailbreaking and Mitigation of Vulnerabilities in Large Language Models","created":"2024-10-20","cited_by_count":34,"area":"safety"},{"id":"2506.17209","title":"Fine-Tuning Lowers Safety and Disrupts Evaluation Consistency","created":"2025-06-20","cited_by_count":27,"area":"safety"},{"id":"2407.20224","title":"Can Editing LLMs Inject Harm?","created":"2024-07-29","cited_by_count":22,"area":"safety"},{"id":"2502.03052","title":"Understanding and Enhancing the Transferability of Jailbreaking Attacks","created":"2025-02-05","cited_by_count":22,"area":"safety"},{"id":"2505.06843","title":"Benign Samples Matter! Fine-tuning On Outlier Benign Samples Severely Breaks Safety","created":"2025-05-11","cited_by_count":22,"area":"safety"},{"id":"2506.07596","title":"TwinBreak: Jailbreaking LLM Security Alignments based on Twin Prompts","created":"2025-06-09","cited_by_count":19,"area":"safety"},{"id":"2408.11006","title":"Security Attacks on LLM-based Code Completion Tools","created":"2024-08-20","cited_by_count":16,"area":"safety"},{"id":"2412.08615","title":"Exploiting the Index Gradients for Optimization-Based Jailbreaking on Large Language Models","created":"2024-12-11","cited_by_count":15,"area":"safety"},{"id":"2410.03857","title":"You Know What I'm Saying: Jailbreak Attack via Implicit Reference","created":"2024-10-04","cited_by_count":14,"area":"safety"},{"id":"2410.03869","title":"Chain-of-Jailbreak Attack for Image Generation Models via Editing Step by Step","created":"2024-10-04","cited_by_count":11,"area":"safety"},{"id":"2505.16559","title":"CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuning","created":"2025-05-22","cited_by_count":11,"area":"safety"},{"id":"2506.06278","title":"Distillation Robustifies Unlearning","created":"2025-06-06","cited_by_count":11,"area":"safety"},{"id":"2505.12186","title":"Self-Destructive Language Model","created":"2025-05-18","cited_by_count":10,"area":"safety"},{"id":"2402.13926","title":"Large Language Models are Vulnerable to Bait-and-Switch Attacks for Generating Harmful Content","created":"2024-02-21","cited_by_count":9,"area":"safety"},{"id":"2507.11630","title":"Jailbreak-Tuning: Models Efficiently Learn Jailbreak Susceptibility","created":"2025-07-15","cited_by_count":9,"area":"safety"},{"id":"2510.07835","title":"MetaDefense: Defending Finetuning-based Jailbreak Attack Before and During Generation","created":"2025-10-09","cited_by_count":8,"area":"safety"},{"id":"2406.18725","title":"Jailbreaking LLMs with Arabic Transliteration and Arabizi","created":"2024-06-26","cited_by_count":7,"area":"safety"},{"id":"2601.10141","title":"Understanding and Preserving Safety in Fine-Tuned LLMs","created":"2026-01-15","cited_by_count":7,"area":"safety"},{"id":"2502.19537","title":"No, of Course I Can! Deeper Fine-Tuning Attacks That Bypass Token-Level Safety Mechanisms","created":"2025-02-26","cited_by_count":6,"area":"safety"},{"id":"2506.12913","title":"Jailbreak Transferability Emerges from Shared Representations","created":"2025-06-15","cited_by_count":6,"area":"safety"},{"id":"2603.00498","title":"Antibody: Strengthening Defense Against Harmful Fine-Tuning for Large Language Models via Attenuating Harmful Gradient Influence","created":"2026-02-28","cited_by_count":6,"area":"safety"},{"id":"2402.05723","title":"In-Context Learning Can Re-learn Forbidden Tasks","created":"2024-02-08","cited_by_count":5,"area":"safety"},{"id":"2407.04151","title":"Securing Multi-turn Conversational Language Models From Distributed Backdoor Triggers","created":"2024-07-04","cited_by_count":5,"area":"safety"},{"id":"2410.09804","title":"BlackDAN: A Black-Box Multi-Objective Approach for Effective and Contextual Jailbreaking of Large Language Models","created":"2024-10-13","cited_by_count":5,"area":"safety"},{"id":"2411.19335","title":"PEFT-as-an-Attack! Jailbreaking Language Models during Federated Parameter-Efficient Fine-Tuning","created":"2024-11-28","cited_by_count":5,"area":"safety"},{"id":"2511.14423","title":"Unified Defense for Large Language Models against Jailbreak and Fine-Tuning Attacks in Education","created":"2025-11-18","cited_by_count":5,"area":"safety"},{"id":"2601.08089","title":"Q-realign: Piggybacking Realignment on Quantization for Safe and Efficient LLM Deployment","created":"2026-01-13","cited_by_count":5,"area":"safety"},{"id":"2602.11495","title":"Jailbreaking Leaves a Trace: Understanding and Detecting Jailbreak Attacks from Internal Representations of Large Language Models","created":"2026-02-12","cited_by_count":4,"area":"safety"},{"id":"2410.02220","title":"Data to Defense: The Role of Curation in Customizing LLMs Against Jailbreaking Attacks","created":"2024-10-03","cited_by_count":3,"area":"safety"},{"id":"2505.04578","title":"Fight Fire with Fire: Defending Against Malicious RL Fine-Tuning via Reward Neutralization","created":"2025-05-07","cited_by_count":3,"area":"safety"},{"id":"2601.03594","title":"Jailbreaking LLMs & VLMs: Mechanisms, Evaluation, and Unified Defense","created":"2026-01-07","cited_by_count":3,"area":"safety"},{"id":"2508.12622","title":"Consiglieres in the Shadow: Understanding the Use of Uncensored Large Language Models in Cybercrimes","created":"2025-08-18","cited_by_count":2,"area":"safety"},{"id":"2510.01342","title":"Fine-Tuning Jailbreaks under Highly Constrained Black-Box Settings: A Three-Pronged Approach","created":"2025-10-01","cited_by_count":2,"area":"safety"},{"id":"2510.10281","title":"ArtPerception: ASCII Art-based Jailbreak on LLMs with Recognition Pre-test","created":"2025-10-11","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":"2607.19424","title":"JailMeter: An Evidence-Based Evaluation Framework for Jailbreak Attacks on Large Language Models","created":"2026-07-20","cited_by_count":2,"area":"safety"},{"id":"2506.18543","title":"SoK: A Comprehensive Security Analysis of Jailbreak Resilience in GPT and DeepSeek Models","created":"2025-06-23","cited_by_count":1,"area":"safety"},{"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":"2603.10068","title":"ADVERSA: Measuring Multi-Turn Guardrail Degradation and Judge Reliability in Large Language Models","created":"2026-03-10","cited_by_count":1,"area":"safety"},{"id":"2605.25073","title":"Security in the Fine-Tuning Lifecycle of Large Language Models: Threats, Defenses,Evaluation, and Future Directions","created":"2026-05-24","cited_by_count":1,"area":"safety"},{"id":"2605.26526","title":"Open-Weight LLM Fine-Tuning Defenses are Susceptible to Simple Attacks","created":"2026-05-26","cited_by_count":1,"area":"safety"},{"id":"2606.12342","title":"ALIGNBEAM : Inference-Time Alignment Transfer via Cross-Vocabulary Logit Mixing","created":"2026-06-10","cited_by_count":1,"area":"safety"},{"id":"2406.11682","title":"Knowledge-to-Jailbreak: Investigating Knowledge-driven Jailbreaking Attacks for Large Language Models","created":"2024-06-17","cited_by_count":0,"area":"safety"},{"id":"2505.23404","title":"AJF: Adaptive Jailbreak Framework Based on the Comprehension Ability of Black-Box Large Language Models","created":"2025-05-29","cited_by_count":0,"area":"safety"},{"id":"2506.00062","title":"SafeCOMM: A Study on Safety Degradation in Fine-Tuned Telecom Large Language Models","created":"2025-05-29","cited_by_count":0,"area":"safety"},{"id":"2511.23174","title":"Are LLMs Good Safety Agents or a Propaganda Engine?","created":"2025-11-28","cited_by_count":0,"area":"safety"},{"id":"2602.04893","title":"A Causal Perspective for Enhancing Jailbreak Attack and Defense","created":"2026-01-31","cited_by_count":0,"area":"safety"},{"id":"2602.17696","title":"Can LLM Safety Be Ensured by Constraining Parameter Regions?","created":"2026-02-06","cited_by_count":0,"area":"safety"},{"id":"2605.21674","title":"Adversarial Reframing: A Framework for Targeted Generation in Language Models","created":"2026-05-20","cited_by_count":0,"area":"safety"},{"id":"2605.22984","title":"Test-Time Training Undermines Safety Guardrails","created":"2026-05-21","cited_by_count":0,"area":"safety"},{"id":"2606.28525","title":"A Gravitational Interpretation of Safety Reversion under Fine-Tuning","created":"2026-06-26","cited_by_count":0,"area":"safety"},{"id":"2606.30263","title":"Defending Against Harmful Supervision Hidden in Benign Samples","created":"2026-06-29","cited_by_count":0,"area":"safety"},{"id":"2610.16305","title":"ATTCK-R: Knowledge-Aware Long-Horizon Jailbreaking for Protected LLMs","created":"2026-10-12","cited_by_count":0,"area":"safety"}],"next_cursor":null}