mlsanityData to 7 Oct 2026

On the Robustness of Machine Unlearning for Vision-Language Models

arXiv 2605.26992 ↗First posted 26 May 2026cs.CVcited by 1

On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness

also in 1 more
Abstract (arXiv)

Vision-language models (VLMs) may memorize undesirable information from training data, motivating growing interest in machine unlearning. In this work, we present the first systematic survey and robustness analysis of VLM unlearning. We provide a comprehensive taxonomy and review of existing VLM unlearning methods, together with unified evaluations under multiple prompt settings. We then propose three attack paradigms to examine whether forgotten multimodal knowledge can be reactivated through contextual prompting or downstream retraining. Extensive experiments show that many existing methods remain vulnerable under these attacks, indicating that current approaches often hide rather than fully remove target knowledge. Our study provides new insights into the robustness and limitations of current VLM unlearning methods and highlights the need for more reliable multimodal unlearning strategies. Code is available at https://github.com/XMUDeepLIT/VLM-UnL-Attack.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 22 of its 34 references that are in the map. First posted on arXiv 26 May 2026.

Builds on

22 of 34 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Towards Benign Memory Forgetting for Selective Multimodal Large Language Model Unlearning25 Nov
  2. 2025MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steering5 Oct
  3. 2025SAUCE: Selective Concept Unlearning in Vision-Language Models with Sparse Autoencoders16 Mar

Other references · 19

  1. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  2. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  3. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,044
  4. 2024LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Modelscited by 1,442
  5. 2017Deeper, Broader and Artier Domain Generalizationcited by 983
  6. 2024TOFU: A Task of Fictitious Unlearning for LLMscited by 507
  7. 2024The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearningcited by 488
  8. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearningcited by 484
  9. 2024Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearningcited by 160
  10. 2025Rethinking LLM Unlearning Objectives: A Gradient Perspective and Go Beyondcited by 59
  11. 2024UNDIAL: Self-Distillation with Adjusted Logits for Robust Unlearning in Large Language Modelscited by 49
  12. 2025Exploring Criteria of Loss Reweighting to Enhance LLM Unlearningcited by 39
7 more
  1. 2022How Large Language Models are Transforming Machine-Paraphrased Plagiarismcited by 38
  2. 2024Supervised Algorithmic Fairness in Distribution Shifts: A Surveycited by 23
  3. 2024Targeted Unlearning with Single Layer Unlearning Gradientcited by 16
  4. 2026Bi-directional Bias Attribution: Debiasing Large Language Models without Modifying Promptscited by 5
  5. 2026Visual-Guided Key-Token Regularization for Multimodal Large Language Model Unlearningcited by 3
  6. 2026Null Space Constrained Contrastive Visual Forgetting for MLLM Unlearningcited by 3
  7. 2026Object Hallucination-Free Reinforcement Unlearning for Vision-Language Modelscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Open-MMUnlearning: Unifying Methods and Evaluation for MLLM Unlearningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Visual-Noise Guided In-Context Distillation for Multimodal Large Language Model Unlearning
  2. 2026One Modality to Forget Them All: Enhancing Cross-Modal Unlearning in Vision-Language Models
  3. 2026Stochastic Meta-Unlearning: Bridging Language Backbone and Multimodal Unlearning
  4. 2026Metric Unreliability in Multimodal Machine Unlearning: A Systematic Analysis and Principled Unified Score
  5. 2026CATA: Continual Machine Unlearning via Conflict-Averse Task Arithmetic
  6. 2026Before Forgetting, Learn to Remember: Revisiting Foundational Learning Failures in LVLM Unlearning Benchmarks
  7. 2025Robust MLLM Unlearning via Visual Knowledge Distillation
  8. 2026Knowledge Vector Weakening: Efficient Training-free Unlearning for Large Vision-Language Models