{"total":63,"rows":[{"id":"2406.14598","title":"SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal","created":"2024-06-20","cited_by_count":240,"area":"alignment"},{"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":"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":"2510.05024","title":"Inoculation Prompting: Instructing LLMs to misbehave at train-time improves test-time alignment","created":"2025-10-06","cited_by_count":43,"area":"alignment"},{"id":"2506.07468","title":"Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Models","created":"2025-06-09","cited_by_count":39,"area":"alignment"},{"id":"2412.19792","title":"InfAlign: Inference-aware language model alignment","created":"2024-12-27","cited_by_count":35,"area":"alignment"},{"id":"2409.14586","title":"Backtracking Improves Generation Safety","created":"2024-09-22","cited_by_count":32,"area":"alignment"},{"id":"2405.20653","title":"Mind the Inconspicuous: Revealing the Hidden Weakness in Aligned LLMs' Refusal Boundaries","created":"2024-05-31","cited_by_count":28,"area":"alignment"},{"id":"2410.15362","title":"Faster-GCG: Efficient Discrete Optimization Jailbreak Attacks against Aligned Large Language Models","created":"2024-10-20","cited_by_count":25,"area":"alignment"},{"id":"2509.11864","title":"NeuroStrike: Neuron-Level Attacks on Aligned LLMs","created":"2025-09-15","cited_by_count":22,"area":"alignment"},{"id":"2602.01539","title":"MAGIC: A Co-Evolving Attacker-Defender Adversarial Game for Robust LLM Safety","created":"2026-02-02","cited_by_count":21,"area":"alignment"},{"id":"2501.14940","title":"CASE-Bench: Context-Aware SafEty Benchmark for Large Language Models","created":"2025-01-24","cited_by_count":20,"area":"alignment"},{"id":"2503.18991","title":"Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment","created":"2025-03-23","cited_by_count":20,"area":"alignment"},{"id":"2509.01909","title":"Oyster-I: Beyond Refusal -- Constructive Safety Alignment for Responsible Language Models","created":"2025-09-02","cited_by_count":19,"area":"alignment"},{"id":"2503.02574","title":"LLM-Safety Evaluations Lack Robustness","created":"2025-03-04","cited_by_count":18,"area":"alignment"},{"id":"2502.04322","title":"Speak Easy: Eliciting Harmful Jailbreaks from LLMs with Simple Interactions","created":"2025-02-06","cited_by_count":15,"area":"alignment"},{"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":"2507.14987","title":"AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning","created":"2025-07-20","cited_by_count":15,"area":"alignment"},{"id":"2505.21605","title":"SoSBench: Benchmarking Safety Alignment on Six Scientific Domains","created":"2025-05-27","cited_by_count":12,"area":"alignment"},{"id":"2406.14393","title":"Jailbreaking as a Reward Misspecification Problem","created":"2024-06-20","cited_by_count":11,"area":"alignment"},{"id":"2510.02833","title":"Attack via Overfitting: 10-shot Benign Fine-tuning to Jailbreak LLMs","created":"2025-10-03","cited_by_count":10,"area":"alignment"},{"id":"2507.11500","title":"ARMOR: Aligning Secure and Safe Large Language Models via Meticulous Reasoning","created":"2025-07-14","cited_by_count":8,"area":"alignment"},{"id":"2505.00038","title":"HyPerAlign: Interpretable Personalized LLM Alignment via Hypothesis Generation","created":"2025-04-29","cited_by_count":7,"area":"alignment"},{"id":"2601.03868","title":"What Matters For Safety Alignment?","created":"2026-01-07","cited_by_count":7,"area":"alignment"},{"id":"2602.06258","title":"GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt","created":"2026-02-05","cited_by_count":7,"area":"alignment"},{"id":"2506.00166","title":"Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignment","created":"2025-05-30","cited_by_count":6,"area":"alignment"},{"id":"2602.07892","title":"Safety Alignment as Continual Learning: Mitigating the Alignment Tax via Orthogonal Gradient Projection","created":"2026-02-08","cited_by_count":6,"area":"alignment"},{"id":"2602.22554","title":"Multilingual Safety Alignment Via Sparse Weight Editing","created":"2026-02-26","cited_by_count":6,"area":"alignment"},{"id":"2507.02332","title":"PII Jailbreaking in LLMs via Activation Steering Reveals Personal Information Leakage","created":"2025-07-03","cited_by_count":5,"area":"alignment"},{"id":"2603.07445","title":"Few Tokens, Big Leverage: Preserving Safety Alignment by Constraining Safety Tokens during Fine-tuning","created":"2026-03-08","cited_by_count":4,"area":"alignment"},{"id":"2606.24014","title":"Reinforcement Learning Towards Broadly and Persistently Beneficial Models","created":"2026-06-22","cited_by_count":4,"area":"alignment"},{"id":"2506.07402","title":"Beyond Jailbreaks: Revealing Stealthier and Broader LLM Security Risks Stemming from Alignment Failures","created":"2025-06-09","cited_by_count":3,"area":"alignment"},{"id":"2602.04448","title":"RASA: Routing-Aware Safety Alignment for Mixture-of-Experts Models","created":"2026-02-04","cited_by_count":3,"area":"alignment"},{"id":"2602.08813","title":"Robust Policy Optimization to Prevent Catastrophic Forgetting","created":"2026-02-09","cited_by_count":3,"area":"alignment"},{"id":"2603.23184","title":"ImplicitRM: Unbiased Reward Modeling from Implicit Preference Data for LLM alignment","created":"2026-03-24","cited_by_count":3,"area":"alignment"},{"id":"2502.11853","title":"StructTransform: A Scalable Attack Surface for Safety-Aligned Large Language Models","created":"2025-02-17","cited_by_count":2,"area":"alignment"},{"id":"2603.22829","title":"Improving Safety Alignment via Balanced Direct Preference Optimization","created":"2026-03-24","cited_by_count":1,"area":"alignment"},{"id":"2604.08844","title":"Spectral Geometry of LoRA Adapters Encodes Training Objective and Predicts Harmful Compliance","created":"2026-04-10","cited_by_count":1,"area":"alignment"},{"id":"2605.24154","title":"Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs","created":"2026-05-22","cited_by_count":1,"area":"alignment"},{"id":"2606.12016","title":"Generalization Hacking: Models Can Game Reinforcement Learning by Preventing Behavioral Generalization","created":"2026-06-10","cited_by_count":1,"area":"alignment"},{"id":"2508.09937","title":"A Comprehensive Evaluation framework of Alignment Techniques for LLMs","created":"2025-08-13","cited_by_count":0,"area":"alignment"},{"id":"2512.05518","title":"Matching Ranks Over Probability Yields Truly Deep Safety Alignment","created":"2025-12-05","cited_by_count":0,"area":"alignment"},{"id":"2602.00619","title":"Jailbreaking LLMs via Calibration","created":"2026-01-31","cited_by_count":0,"area":"alignment"},{"id":"2604.02574","title":"Understanding the Effects of Safety Unalignment on Large Language Models","created":"2026-04-02","cited_by_count":0,"area":"alignment"},{"id":"2605.08878","title":"Why Do Aligned LLMs Remain Jailbreakable: Refusal-Escape Directions, Operator-Level Sources, and Safety-Utility Trade-off","created":"2026-05-09","cited_by_count":0,"area":"alignment"},{"id":"2605.10998","title":"Few-Shot Truly Benign DPO Attack for Jailbreaking LLMs","created":"2026-05-09","cited_by_count":0,"area":"alignment"},{"id":"2605.20654","title":"REFLECTOR: Internalizing Step-wise Reflection against Indirect Jailbreak","created":"2026-05-20","cited_by_count":0,"area":"alignment"},{"id":"2605.21362","title":"LASH: Adaptive Semantic Hybridization for Black-Box Jailbreaking of Large Language Models","created":"2026-05-20","cited_by_count":0,"area":"alignment"},{"id":"2606.00651","title":"MESA: Improving MoE Safety Alignment via Decentralized Expertise","created":"2026-05-30","cited_by_count":0,"area":"alignment"},{"id":"2606.16349","title":"From Refusal Geometry to Safety Geometry: Harmfulness--Refusal Coupling under Dynamic Adversarial Fine-Tuning","created":"2026-06-15","cited_by_count":0,"area":"alignment"},{"id":"2607.02914","title":"Oyster-II: Reinforcement Learning for Constructive Safety Alignment in Large Language Models","created":"2026-07-03","cited_by_count":0,"area":"alignment"},{"id":"2607.03453","title":"Best-of-Better-$N$: Generating Pre-Aligned Responses with In-Context Learning","created":"2026-07-03","cited_by_count":0,"area":"alignment"},{"id":"2607.16242","title":"TRACE: Trajectory-Based Safety Patch Learning for LLM Post-Training Realignment","created":"2026-06-26","cited_by_count":0,"area":"alignment"},{"id":"2607.20494","title":"Isolating LLM Alignment from Regex: Zero Coverage and Metric-Dependent Divergence Under Adversarial Mutation","created":"2026-06-12","cited_by_count":0,"area":"alignment"},{"id":"2608.01414","title":"No Single Neuron of Failure: Distributed Safety Alignment Against White-Box Attacks","created":"2026-08-02","cited_by_count":0,"area":"alignment"},{"id":"2608.11705","title":"Making Your LLMs More Objective: Stabilizing LLM Safety Behavior Across Traits with Trait-Invariant Safety Tuning","created":"2026-08-12","cited_by_count":0,"area":"alignment"},{"id":"2608.17202","title":"Fool's Gold: Defensive Deception Against Safety-Removal Attacks on Open-Weight Models","created":"2026-08-17","cited_by_count":0,"area":"alignment"},{"id":"2609.06934","title":"The Geometry of Refusal: Why Post-Hoc Safety Is Fragile and Pretraining-Time Safety Persists","created":"2026-09-07","cited_by_count":0,"area":"alignment"},{"id":"2609.34033","title":"Faithful Activation Verbalization: Reducing Hallucinations in LLM Representation Interpretation","created":"2026-09-27","cited_by_count":0,"area":"alignment"},{"id":"2609.34519","title":"EOPSA: Efficient On-Policy Self-Distilled Safety Alignment","created":"2026-09-28","cited_by_count":0,"area":"alignment"},{"id":"2609.34771","title":"When Do Model Internals Help? Exploring the Role of Representation Engineering in LLM Safety","created":"2026-09-28","cited_by_count":0,"area":"alignment"},{"id":"2609.38389","title":"The Geometry of Harmfulness in Multi-Turn Attacks","created":"2026-09-29","cited_by_count":0,"area":"alignment"},{"id":"2609.39902","title":"CodeMimicry: Exploiting Safety Generalization Lag in Large Language Models via Structured Code Completion","created":"2026-09-30","cited_by_count":0,"area":"alignment"}],"next_cursor":null}