Decoupling Knowledge Forgetting and Refusal Learning in MLLM Unlearning
cs.AI, cs.CL, cs.IR, cs.LG
Submitted: 2026-10-10
Updated: 2026-10-10
Terminology
Sources
- Qwen3-VL Technical Report
- SafeEraser: Enhancing Safety in Multimodal Large Language Models through Multimodal Machine Unlearning
- MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steering
- Reducing the Safety Tax in LLM Safety Alignment with On-Policy Self-Distillation
- Unlearners Can Lie: Evaluating and Improving Honesty in LLM Unlearning
- ASRU: Activation Steering Meets Reinforcement Unlearning for Multimodal Large Language Models
- Distilling the Knowledge in a Neural Network
- Language Models (Mostly) Know What They Know
- Sequence-Level Knowledge Distillation
- DistiLLM: Towards Streamlined Distillation for Large Language Models
- Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
- Improved Baselines with Visual Instruction Tuning
- Towards Safer Large Language Models through Machine Unlearning
- X-Boundary: Establishing Exact Safety Boundary to Shield LLMs from Multi-Turn Jailbreaks without Compromising Usability
- Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset
- TOFU: A Task of Fictitious Unlearning for LLMs
- A Reduction of Imitation Learning and Structured Prediction to No-Regret Online Learning
- Self-Distillation Enables Continual Learning
- AlphaSteer: Learning Refusal Steering with Principled Null-Space Constraint
- A Survey of On-Policy Distillation for Large Language Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems