{"total":70,"rows":[{"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":"2502.05163","title":"Enhancing LLM Safety Through a Theoretical Minimax Game Lens","created":"2025-02-07","cited_by_count":19,"area":"safety"},{"id":"2603.13026","title":"PISmith: Reinforcement Learning-based Red Teaming for Prompt Injection Defenses","created":"2026-03-13","cited_by_count":16,"area":"safety"},{"id":"2512.07761","title":"TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards","created":"2025-12-08","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":"2602.06854","title":"SEMA: Simple yet Effective Learning for Multi-Turn Jailbreak Attacks","created":"2026-02-06","cited_by_count":12,"area":"safety"},{"id":"2502.12658","title":"R.R.: Unveiling LLM Training Privacy through Recollection and Ranking","created":"2025-02-18","cited_by_count":10,"area":"safety"},{"id":"2506.07121","title":"Quality-Diversity Red-Teaming: Automated Generation of High-Quality and Diverse Attackers for Large Language Models","created":"2025-06-08","cited_by_count":9,"area":"safety"},{"id":"2510.02286","title":"Tree-based Dialogue Reinforced Policy Optimization for Red-Teaming Attacks","created":"2025-10-02","cited_by_count":8,"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":"2510.16629","title":"On the Impossibility of Retrain Equivalence in Machine Unlearning","created":"2025-10-18","cited_by_count":6,"area":"safety"},{"id":"2603.01228","title":"Towards Policy-Adaptive Image Guardrail: Benchmark and Method","created":"2026-03-01","cited_by_count":6,"area":"safety"},{"id":"2604.01194","title":"AgentWatcher: A Rule-based Prompt Injection Monitor","created":"2026-04-01","cited_by_count":6,"area":"safety"},{"id":"2505.15656","title":"Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!","created":"2025-05-21","cited_by_count":5,"area":"safety"},{"id":"2505.16567","title":"Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM Finetuning","created":"2025-05-22","cited_by_count":5,"area":"safety"},{"id":"2510.07239","title":"Red-Bandit: Test-Time Adaptation for LLM Red-Teaming via Bandit-Guided LoRA Experts","created":"2025-10-08","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":"2503.01224","title":"CE-U: Cross Entropy Unlearning","created":"2025-03-03","cited_by_count":4,"area":"safety"},{"id":"2601.20568","title":"Reinforcement Unlearning via Group Relative Policy Optimization","created":"2026-01-28","cited_by_count":4,"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":"2510.08158","title":"Beyond Over-Refusal: Scenario-Based Diagnostics and Post-Hoc Mitigation for Exaggerated Refusals in LLMs","created":"2025-10-09","cited_by_count":3,"area":"safety"},{"id":"2605.08778","title":"Not All Turns Matter: Credit Assignment for Multi-Turn Jailbreaking","created":"2026-05-09","cited_by_count":3,"area":"safety"},{"id":"2505.16643","title":"From Evaluation to Defense: Advancing Safety in Video Large Language Models","created":"2025-05-22","cited_by_count":2,"area":"safety"},{"id":"2508.04281","title":"Prompt Injection Vulnerability of Consensus Generating Applications in Digital Democracy","created":"2025-08-06","cited_by_count":2,"area":"safety"},{"id":"2508.09190","title":"Multi-Level Safety Continual Projection for Fine-Tuned Large Language Models without Retraining","created":"2025-08-08","cited_by_count":2,"area":"safety"},{"id":"2509.20924","title":"RLCracker: Evaluating the Worst-Case Vulnerability of LLM Watermarks with Adaptive RL Attacks","created":"2025-09-25","cited_by_count":2,"area":"safety"},{"id":"2510.05169","title":"From Poisoned to Aware: Fostering Backdoor Self-Awareness in LLMs","created":"2025-10-05","cited_by_count":2,"area":"safety"},{"id":"2603.02436","title":"TraceGuard: Process-Guided Firewall against Reasoning Backdoors in Large Language Models","created":"2026-03-02","cited_by_count":2,"area":"safety"},{"id":"2604.11546","title":"RLSpoofer: A Lightweight Evaluator for LLM Watermark Spoofing Resilience","created":"2026-04-13","cited_by_count":2,"area":"safety"},{"id":"2604.22438","title":"SSG: Logit-Balanced Vocabulary Partitioning for LLM Watermarking","created":"2026-04-24","cited_by_count":2,"area":"safety"},{"id":"2506.04302","title":"RedRFT: A Light-Weight Benchmark for Reinforcement Fine-Tuning-Based Red Teaming","created":"2025-06-04","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":"2508.11925","title":"Optimizing Token Choice for Code Watermarking: An RL Approach","created":"2025-08-16","cited_by_count":1,"area":"safety"},{"id":"2511.05867","title":"Can LLM Infer Risk Information From MCP Server System Logs?","created":"2025-11-08","cited_by_count":1,"area":"safety"},{"id":"2603.29038","title":"Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning","created":"2026-03-30","cited_by_count":1,"area":"safety"},{"id":"2605.00553","title":"Stable-GFlowNet: Toward Diverse and Robust LLM Red-Teaming via Contrastive Trajectory Balance","created":"2026-05-01","cited_by_count":1,"area":"safety"},{"id":"2606.27210","title":"Paved with True Intents: Intent-Aware Training Improves LLM Safety Classification Across Training Regimes","created":"2026-06-25","cited_by_count":1,"area":"safety"},{"id":"2607.18056","title":"An Early Warning of Emerging Biosecurity Risks in Frontier LLMs","created":"2026-07-20","cited_by_count":1,"area":"safety"},{"id":"2607.27968","title":"Beyond Binary Rewards: A Comparative Study of Reward Design for Reinforcement Unlearning","created":"2026-07-30","cited_by_count":1,"area":"safety"},{"id":"2608.23028","title":"PsychJail: Exploring Psychological Jailbreaks via Multi-Turn Persuasion of LLM Policies","created":"2026-08-24","cited_by_count":1,"area":"safety"},{"id":"2508.15648","title":"SDGO: Self-Discrimination-Guided Optimization for Consistent Safety in Large Language Models","created":"2025-08-21","cited_by_count":0,"area":"safety"},{"id":"2511.04215","title":"Black-Box Guardrail Reverse-engineering Attack","created":"2025-11-06","cited_by_count":0,"area":"safety"},{"id":"2512.04044","title":"MarkTune: Improving the Quality-Detectability Trade-off in Open-Weight LLM Watermarking","created":"2025-12-03","cited_by_count":0,"area":"safety"},{"id":"2602.02536","title":"From Sparse Decisions to Dense Reasoning: A Multi-attribute Trajectory Paradigm for Multimodal Moderation","created":"2026-01-28","cited_by_count":0,"area":"safety"},{"id":"2602.04196","title":"The Missing Half: Unveiling Training-time Implicit Safety Risks Beyond Deployment","created":"2026-02-04","cited_by_count":0,"area":"safety"},{"id":"2602.12124","title":"Alignment Risks from Capability-Seeking RL Training","created":"2026-02-12","cited_by_count":0,"area":"safety"},{"id":"2602.19020","title":"Learning to Detect Language Model Training Data via Active Reconstruction","created":"2026-02-22","cited_by_count":0,"area":"safety"},{"id":"2604.05716","title":"Can Large Language Models Reinvent Foundational Algorithms?","created":"2026-04-07","cited_by_count":0,"area":"safety"},{"id":"2604.18510","title":"Behind Harmful Compliance: Behavioral and Mechanistic Divergence Across LLM Jailbreaks","created":"2026-04-20","cited_by_count":0,"area":"safety"},{"id":"2604.25247","title":"R-CoT: A Reasoning-Layer Watermark via Redundant Chain-of-Thought in Large Language Models","created":"2026-04-28","cited_by_count":0,"area":"safety"},{"id":"2605.00699","title":"STARE: Step-wise Temporal Alignment and Red-teaming Engine for Multi-modal Toxicity Attack","created":"2026-05-01","cited_by_count":0,"area":"safety"},{"id":"2605.13411","title":"Model-Agnostic Lifelong LLM Safety via Externalized Attack-Defense Co-Evolution","created":"2026-05-13","cited_by_count":0,"area":"safety"},{"id":"2605.30919","title":"De-attribute to Forget for LLM Unlearning","created":"2026-05-29","cited_by_count":0,"area":"safety"},{"id":"2606.05614","title":"Safety Paradox: How Enhanced Safety Awareness Leaves LLMs Vulnerable to Posterior Attack","created":"2026-06-04","cited_by_count":0,"area":"safety"},{"id":"2606.15333","title":"Replay What Matters: Off-Policy Replay for Efficient LLM Reinforcement Unlearning","created":"2026-06-13","cited_by_count":0,"area":"safety"},{"id":"2606.19222","title":"Mechanism-Guided Selective Unlearning for RLVR-Induced Reasoning","created":"2026-06-17","cited_by_count":0,"area":"safety"},{"id":"2606.27632","title":"Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety","created":"2026-06-26","cited_by_count":0,"area":"safety"},{"id":"2607.00361","title":"ReShift: Aha-Moment-Driven Reasoning-Level Backdoor Attacks on Vision-Language Models","created":"2026-07-01","cited_by_count":0,"area":"safety"},{"id":"2607.11070","title":"MJ: Multi-turn LLM Jailbreaking via Decomposed Credit Assignment","created":"2026-07-13","cited_by_count":0,"area":"safety"},{"id":"2607.20472","title":"Robust Critics: Defending LLMs Against Multi-Turn Attacks","created":"2026-05-24","cited_by_count":0,"area":"safety"},{"id":"2608.12346","title":"Position: The Alignment Community is Unintentionally Building a Censor's Toolkit","created":"2026-06-05","cited_by_count":0,"area":"safety"},{"id":"2608.29030","title":"Learning to Follow In-Context Watermark Instructions via Self-Distillation","created":"2026-08-29","cited_by_count":0,"area":"safety"},{"id":"2609.29948","title":"ENDOPROMPT: Victim-Side Pseudo-References for Utility Degradation","created":"2026-09-24","cited_by_count":0,"area":"safety"},{"id":"2609.32801","title":"PlanGuard: A Guardrail for Multi-Step Plan Safety in Embodied Agents","created":"2026-09-26","cited_by_count":0,"area":"safety"},{"id":"2609.33628","title":"Climbing the Hill: Prompt Injection Red-Teaming Against Frontier Models with Curriculum Reinforcement Learning","created":"2026-09-27","cited_by_count":0,"area":"safety"},{"id":"2609.36121","title":"Render Before Reading: Visual Rendering as a Prompt Injection Defense","created":"2026-09-28","cited_by_count":0,"area":"safety"},{"id":"2610.05282","title":"Red-TTT: Test-Time Training for Automated Jailbreaking Large Language Models","created":"2026-10-04","cited_by_count":0,"area":"safety"},{"id":"2610.05346","title":"Reflections and Fragments: Securing LLMs Against Sequential Mosaic Attacks","created":"2026-10-04","cited_by_count":0,"area":"safety"},{"id":"2610.14701","title":"Towards Merging-conditional Backdoor Unalignment against Large Language Models","created":"2026-10-10","cited_by_count":0,"area":"safety"},{"id":"2610.19133","title":"TRACER: Trace-Regularized Training Against Indirect Prompt Injection","created":"2026-10-13","cited_by_count":0,"area":"safety"}],"next_cursor":null}