{"total":139,"rows":[{"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":"2408.15221","title":"LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet","created":"2024-08-27","cited_by_count":151,"area":"safety"},{"id":"2507.11878","title":"LLMs Encode Harmfulness and Refusal Separately","created":"2025-07-16","cited_by_count":88,"area":"safety"},{"id":"2502.05374","title":"Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond","created":"2025-02-07","cited_by_count":79,"area":"safety"},{"id":"2406.18682","title":"The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm","created":"2024-06-26","cited_by_count":70,"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":"2506.07022","title":"AlphaSteer: Learning Refusal Steering with Principled Null-Space Constraint","created":"2025-06-08","cited_by_count":49,"area":"safety"},{"id":"2505.11953","title":"Exploring Criteria of Loss Reweighting to Enhance LLM Unlearning","created":"2025-05-17","cited_by_count":39,"area":"safety"},{"id":"2504.01550","title":"Representation Bending for Large Language Model Safety","created":"2025-04-02","cited_by_count":30,"area":"safety"},{"id":"2412.10321","title":"AdvPrefix: An Objective for Nuanced LLM Jailbreaks","created":"2024-12-13","cited_by_count":25,"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":"2503.00187","title":"Steering Dialogue Dynamics for Robustness against Multi-turn Jailbreaking Attacks","created":"2025-02-28","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":"2508.07173","title":"Omni-SafetyBench: A Benchmark for Safety Evaluation of Audio-Visual Large Language Models","created":"2025-08-10","cited_by_count":18,"area":"safety"},{"id":"2505.15710","title":"Advancing LLM Safe Alignment with Safety Representation Ranking","created":"2025-05-21","cited_by_count":16,"area":"safety"},{"id":"2510.27062","title":"Consistency Training Helps Stop Sycophancy and Jailbreaks","created":"2025-10-31","cited_by_count":16,"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":"2503.08195","title":"Dialogue Injection Attack: Jailbreaking LLMs through Context Manipulation","created":"2025-03-11","cited_by_count":14,"area":"safety"},{"id":"2505.19056","title":"An Embarrassingly Simple Defense Against LLM Abliteration Attacks","created":"2025-05-25","cited_by_count":13,"area":"safety"},{"id":"2507.07341","title":"On the Impossibility of Separating Intelligence from Judgment: The Computational Intractability of Filtering for AI Alignment","created":"2025-07-09","cited_by_count":12,"area":"safety"},{"id":"2509.24296","title":"DiffuGuard: How Intrinsic Safety is Lost and Found in Diffusion Large Language Models","created":"2025-09-29","cited_by_count":12,"area":"safety"},{"id":"2602.06854","title":"SEMA: Simple yet Effective Learning for Multi-Turn Jailbreak Attacks","created":"2026-02-06","cited_by_count":12,"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":"2506.06112","title":"Towards Lifecycle Unlearning Commitment Management: Measuring Sample-level Unlearning Completeness","created":"2025-06-06","cited_by_count":11,"area":"safety"},{"id":"2508.12398","title":"Where to Start Alignment? Diffusion Large Language Model May Demand a Distinct Position","created":"2025-08-17","cited_by_count":11,"area":"safety"},{"id":"2408.10668","title":"Probing the Safety Response Boundary of Large Language Models via Unsafe Decoding Path Generation","created":"2024-08-20","cited_by_count":10,"area":"safety"},{"id":"2504.21038","title":"Prefill-level Jailbreak: A Black-Box Risk Analysis of Large Language Models","created":"2025-04-28","cited_by_count":10,"area":"safety"},{"id":"2505.12186","title":"Self-Destructive Language Model","created":"2025-05-18","cited_by_count":10,"area":"safety"},{"id":"2505.17601","title":"Revisiting Backdoor Attacks on LLMs: A Stealthy and Practical Poisoning Framework via Harmless Inputs","created":"2025-05-23","cited_by_count":10,"area":"safety"},{"id":"2506.12880","title":"Universal Jailbreak Suffixes Are Strong Attention Hijackers","created":"2025-06-15","cited_by_count":10,"area":"safety"},{"id":"2603.24543","title":"Analysing the Safety Pitfalls of Steering Vectors","created":"2026-03-25","cited_by_count":10,"area":"safety"},{"id":"2409.07503","title":"AdaPPA: Adaptive Position Pre-Fill Jailbreak Attack Approach Targeting LLMs","created":"2024-09-11","cited_by_count":8,"area":"safety"},{"id":"2510.17633","title":"SARSteer: Safeguarding Large Audio-Language Models via Safe-Ablated Refusal Steering","created":"2025-10-20","cited_by_count":8,"area":"safety"},{"id":"2601.09172","title":"BalDRO: A Distributionally Robust Optimization based Framework for Large Language Model Unlearning","created":"2026-01-14","cited_by_count":8,"area":"safety"},{"id":"2605.01687","title":"MultiBreak: A Scalable and Diverse Multi-turn Jailbreak Benchmark for Evaluating LLM Safety","created":"2026-05-03","cited_by_count":8,"area":"safety"},{"id":"2601.07087","title":"When Should We Introduce Safety Interventions During Pretraining?","created":"2026-01-11","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":"2501.13677","title":"HumorReject: Decoupling LLM Safety from Refusal Prefix via A Little Humor","created":"2025-01-23","cited_by_count":6,"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":"2504.05652","title":"Sugar-Coated Poison: Benign Generation Unlocks LLM Jailbreaking","created":"2025-04-08","cited_by_count":6,"area":"safety"},{"id":"2506.07452","title":"When Style Breaks Safety: Defending LLMs Against Superficial Style Alignment","created":"2025-06-09","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":"2508.03936","title":"ASTRA: Autonomous Spatial-Temporal Red-teaming for AI Software Assistants","created":"2025-08-05","cited_by_count":6,"area":"safety"},{"id":"2510.17210","title":"Wisdom is Knowing What not to Say: Hallucination-Free LLMs Unlearning via Attention Shifting","created":"2025-10-20","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":"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":"2503.11185","title":"Bleeding Pathways: Vanishing Discriminability in LLM Hidden States Fuels Jailbreak Attacks","created":"2025-03-14","cited_by_count":5,"area":"safety"},{"id":"2505.17519","title":"Chain-of-Lure: A Universal Jailbreak Attack Framework using Unconstrained Synthetic Narratives","created":"2025-05-23","cited_by_count":5,"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":"2509.10931","title":"Harmful Prompt Laundering: Jailbreaking LLMs with Abductive Styles and Symbolic Encoding","created":"2025-09-13","cited_by_count":5,"area":"safety"},{"id":"2509.22263","title":"Erase or Hide? Suppressing Spurious Unlearning Neurons for Robust Unlearning","created":"2025-09-26","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":"2502.16174","title":"Efficient LLM Moderation with Multi-Layer Latent Prototypes","created":"2025-02-22","cited_by_count":4,"area":"safety"},{"id":"2503.08919","title":"Backtracking for Safety","created":"2025-03-11","cited_by_count":4,"area":"safety"},{"id":"2503.24191","title":"When Grammar Guides the Attack: Uncovering Control-Plane Vulnerabilities in LLMs with Structured Output","created":"2025-03-31","cited_by_count":4,"area":"safety"},{"id":"2506.17125","title":"Large Language Model Unlearning for Source Code","created":"2025-06-20","cited_by_count":4,"area":"safety"},{"id":"2506.17299","title":"Toward Principled LLM Safety Testing: Solving the Jailbreak Oracle Problem","created":"2025-06-17","cited_by_count":4,"area":"safety"},{"id":"2508.12072","title":"Mitigating Jailbreaks with Intent-Aware LLMs","created":"2025-08-16","cited_by_count":4,"area":"safety"},{"id":"2601.04740","title":"StealthGraph: Exposing Domain-Specific Risks in LLMs through Knowledge-Graph-Guided Harmful Prompt Generation","created":"2026-01-08","cited_by_count":4,"area":"safety"},{"id":"2604.08169","title":"Activation Steering for Aligned Open-ended Generation without Sacrificing Coherence","created":"2026-04-09","cited_by_count":4,"area":"safety"},{"id":"2501.16534","title":"Targeting Alignment: Extracting Safety Classifiers of Aligned LLMs","created":"2025-01-27","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":"2510.02422","title":"Dynamic Jailbreaking Attack","created":"2025-10-02","cited_by_count":3,"area":"safety"},{"id":"2601.15111","title":"Auditing Language Model Unlearning via Information Decomposition","created":"2026-01-21","cited_by_count":3,"area":"safety"},{"id":"2602.13547","title":"AISA: Awakening Intrinsic Safety Awareness in Large Language Models against Jailbreak Attacks","created":"2026-02-14","cited_by_count":3,"area":"safety"},{"id":"2604.09544","title":"Large Language Models Generate Harmful Responses Using a Distinct Mechanism, Shared Across Harm Types","created":"2026-04-10","cited_by_count":3,"area":"safety"},{"id":"2605.05058","title":"SoK: Robustness in Large Language Models against Jailbreak Attacks","created":"2026-05-06","cited_by_count":3,"area":"safety"},{"id":"2412.12192","title":"No Free Lunch for Defending Against Prefilling Attack by In-Context Learning","created":"2024-12-13","cited_by_count":2,"area":"safety"},{"id":"2506.05376","title":"A Red Teaming Roadmap Towards System-Level Safety","created":"2025-05-30","cited_by_count":2,"area":"safety"},{"id":"2506.17881","title":"GRAF: Multi-turn Jailbreaking via Global Refinement and Active Fabrication","created":"2025-06-22","cited_by_count":2,"area":"safety"},{"id":"2508.16484","title":"HAMSA: Hijacking Aligned Compact Models via Stealthy Automation","created":"2025-08-22","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":"2511.01375","title":"Align to Misalign: Automatic LLM Jailbreak with Meta-Optimized LLM Judges","created":"2025-11-03","cited_by_count":2,"area":"safety"},{"id":"2602.19275","title":"KUDA: Knowledge Unlearning by Deviating Representation for Large Language Models","created":"2026-02-22","cited_by_count":2,"area":"safety"},{"id":"2604.08297","title":"Towards Identification and Intervention of Safety-Critical Parameters in Large Language Models","created":"2026-04-09","cited_by_count":2,"area":"safety"},{"id":"2605.04700","title":"Sparse Tokens Suffice: Jailbreaking Audio Language Models via Token-Aware Gradient Optimization","created":"2026-05-06","cited_by_count":2,"area":"safety"},{"id":"2605.31293","title":"Divergence Decoding: Inference-Time Unlearning via Auxiliary Models","created":"2026-05-29","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":"2607.19806","title":"OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization","created":"2026-07-22","cited_by_count":2,"area":"safety"},{"id":"2503.03502","title":"Geometry-Guided Adversarial Prompt Detection via Curvature and Local Intrinsic Dimension","created":"2025-03-05","cited_by_count":1,"area":"safety"},{"id":"2506.01770","title":"ReGA: Model-Based Safeguard for LLMs via Representation-Guided Abstraction","created":"2025-06-02","cited_by_count":1,"area":"safety"},{"id":"2507.01752","title":"Tuning without Peeking: Provable Generalization Bounds and Robust LLM Post-Training","created":"2025-07-02","cited_by_count":1,"area":"safety"},{"id":"2507.22160","title":"Strategic Deflection: Defending LLMs from Logit Manipulation","created":"2025-07-29","cited_by_count":1,"area":"safety"},{"id":"2508.00555","title":"Activation-Guided Local Editing for Jailbreaking Attacks","created":"2025-08-01","cited_by_count":1,"area":"safety"},{"id":"2509.25843","title":"ASGuard: Activation-Scaling Guard to Mitigate Targeted Jailbreaking Attack","created":"2025-09-30","cited_by_count":1,"area":"safety"},{"id":"2511.08367","title":"Why Does Weak-OOD Help? A Further Step Towards Understanding Jailbreaking VLMs","created":"2025-11-11","cited_by_count":1,"area":"safety"},{"id":"2512.10998","title":"SCOUT: A Defense Against Data Poisoning Attacks in Fine-Tuned Language Models","created":"2025-12-10","cited_by_count":1,"area":"safety"},{"id":"2601.12460","title":"TrojanPraise: Jailbreak LLMs via Benign Fine-Tuning","created":"2026-01-18","cited_by_count":1,"area":"safety"},{"id":"2602.06623","title":"Do Prompts Guarantee Safety? Mitigating Toxicity from LLM Generations through Subspace Intervention","created":"2026-02-06","cited_by_count":1,"area":"safety"},{"id":"2602.07107","title":"ShallowJail: Steering Jailbreaks against Large Language Models","created":"2026-02-06","cited_by_count":1,"area":"safety"},{"id":"2603.08234","title":"The Struggle Between Continuation and Refusal: A Mechanistic Analysis of the Continuation-Triggered Jailbreak in LLMs","created":"2026-03-09","cited_by_count":1,"area":"safety"},{"id":"2604.01444","title":"Cooking Up Risks: Benchmarking and Reducing Food Safety Risks in Large Language Models","created":"2026-04-01","cited_by_count":1,"area":"safety"},{"id":"2604.15780","title":"Pruning Unsafe Tickets: A Resource-Efficient Framework for Safer and More Robust LLMs","created":"2026-04-17","cited_by_count":1,"area":"safety"},{"id":"2604.25921","title":"One Word at a Time: Incremental Completion Decomposition Breaks LLM Safety","created":"2026-04-01","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":"2606.04027","title":"MaskForge: Structure-Aware Adaptive Attacks for Jailbreaking Diffusion Large Language Models","created":"2026-06-01","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":"2607.27081","title":"On-Policy Distillation for LLM Safety: A Routing Approach to Template-Robust Realignment","created":"2026-07-29","cited_by_count":1,"area":"safety"},{"id":"2608.03201","title":"When Refusal Looks Safe: The Refusal-Cue Shortcut in Safety Guard Models","created":"2026-08-04","cited_by_count":1,"area":"safety"}],"next_cursor":"100"}