{"total":34,"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":"2502.07557","title":"JBShield: Defending Large Language Models from Jailbreak Attacks through Activated Concept Analysis and Manipulation","created":"2025-02-11","cited_by_count":54,"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":"2502.07218","title":"LLM Unlearning via Neural Activation Redirection","created":"2025-02-11","cited_by_count":41,"area":"safety"},{"id":"2406.12094","title":"Who's asking? User personas and the mechanics of latent misalignment","created":"2024-06-17","cited_by_count":40,"area":"safety"},{"id":"2411.03343","title":"What Features in Prompts Jailbreak LLMs? Investigating the Mechanisms Behind Attacks","created":"2024-11-02","cited_by_count":29,"area":"safety"},{"id":"2411.11114","title":"JailbreakLens: Interpreting Jailbreak Mechanism in the Lens of Representation and Circuit","created":"2024-11-17","cited_by_count":22,"area":"safety"},{"id":"2509.26238","title":"Beyond Linear Probes: Dynamic Safety Monitoring for Language Models","created":"2025-09-30","cited_by_count":16,"area":"safety"},{"id":"2505.24445","title":"Learning Safety Constraints for Large Language Models","created":"2025-05-30","cited_by_count":13,"area":"safety"},{"id":"2506.07891","title":"Video Unlearning via Low-Rank Refusal Vector","created":"2025-06-09","cited_by_count":13,"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":"2602.00298","title":"Assessing Domain-Level Susceptibility to Emergent Misalignment from Narrow Finetuning","created":"2026-01-30","cited_by_count":3,"area":"safety"},{"id":"2603.27412","title":"The Geometry of Harmful Intent: Training-Free Anomaly Detection via Angular Deviation in LLM Residual Streams","created":"2026-03-28","cited_by_count":3,"area":"safety"},{"id":"2601.21564","title":"Representation Unlearning: Forgetting through Information Compression","created":"2026-01-29","cited_by_count":2,"area":"safety"},{"id":"2605.05443","title":"SLAM: Structural Linguistic Activation Marking for Language Models","created":"2026-05-06","cited_by_count":2,"area":"safety"},{"id":"2605.12765","title":"Inference-Time Machine Unlearning via Gated Activation Redirection","created":"2026-05-12","cited_by_count":2,"area":"safety"},{"id":"2503.09095","title":"Backdooring CLIP through Concept Confusion","created":"2025-03-12","cited_by_count":1,"area":"safety"},{"id":"2511.19009","title":"Understanding and Mitigating Over-refusal for Large Language Models via Representation Intervention","created":"2025-11-24","cited_by_count":1,"area":"safety"},{"id":"2601.21702","title":"Beyond Forgetting: Representation Misdirection Elicits Controllable Side Behaviors and Capabilities","created":"2026-01-29","cited_by_count":1,"area":"safety"},{"id":"2603.10806","title":"Backdoor Directions in Vision Transformers","created":"2026-03-11","cited_by_count":1,"area":"safety"},{"id":"2505.18672","title":"Concept Concentration for Faithful Representation Intervention","created":"2025-05-24","cited_by_count":0,"area":"safety"},{"id":"2605.00123","title":"Minimal, Local, Causal Explanations for Jailbreak Success in Large Language Models","created":"2026-04-30","cited_by_count":0,"area":"safety"},{"id":"2605.07032","title":"A Systematic Investigation of RL-Jailbreaking in LLMs","created":"2026-05-07","cited_by_count":0,"area":"safety"},{"id":"2605.10633","title":"Intrinsic Guardrails: How Semantic Geometry of Personality Interacts with Emergent Misalignment in LLMs","created":"2026-05-11","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":"2607.15218","title":"When Words Are Safe But Actions Kill: Probing Physical Jailbreak Beyond Textual Jailbreak in Hidden-State Risk Space","created":"2026-07-16","cited_by_count":0,"area":"safety"},{"id":"2607.19366","title":"Geometry-Guided Constraint Learning for LLM Safety Classification","created":"2026-06-09","cited_by_count":0,"area":"safety"},{"id":"2609.13534","title":"Harmfulness Propagation Dynamics: Layer-wise Trajectories of Adversarial Intent in Large Language Models","created":"2026-09-11","cited_by_count":0,"area":"safety"},{"id":"2609.20129","title":"Local Sparsity Enables Unsupervised LLM Safety Detection","created":"2026-09-17","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":"2609.38929","title":"Learning What to Forget: Distributional Unlearning for LLM Representation Spaces","created":"2026-09-30","cited_by_count":0,"area":"safety"},{"id":"2609.39573","title":"Steering Fields: Adaptive Vector Fields for Safe Image Generation and Beyond","created":"2026-09-30","cited_by_count":0,"area":"safety"},{"id":"2610.04245","title":"On the Steering Dimensionality of Refusal in Language Models","created":"2026-10-03","cited_by_count":0,"area":"safety"},{"id":"2610.06064","title":"TrustMI: Causally controlling how assistants trust their users","created":"2026-10-05","cited_by_count":0,"area":"safety"}],"next_cursor":null}