mlsanityData to 14 Oct 2026

MMUnlearner: Reformulating Multimodal Machine Unlearning in the Era of Multimodal Large Language Models

arXiv 2502.11051 ↗First posted 16 Feb 2025cs.CL cs.AIcited by 30

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

also in 1 more
Abstract (arXiv)

Recent progress in Machine Unlearning (MU) has introduced solutions for the selective removal of private or sensitive information encoded within deep neural networks. Nonetheless, MU for Multimodal Large Language Models (MLLMs) remains in its nascent phase. Therefore, we propose to reformulate the task of multimodal MU in the era of MLLMs, which aims to erase only the visual patterns associated with a given entity while preserving the corresponding textual knowledge encoded within the original parameters of the language model backbone. Furthermore, we develop a novel geometry-constrained gradient ascent method MMUnlearner. It updates the weights of MLLMs with a weight saliency map jointly restricted by the remaining concepts and textual knowledge during unlearning, thereby preserving parameters essential for non-target knowledge. Extensive experiments demonstrate that MMUnlearner surpasses baselines that finetuning MLLMs with VQA data directly through Gradient Ascent (GA) or Negative Preference Optimization (NPO), across all evaluation dimensions. Our code can be found in [this URL](https://github.com/Z1zs/MMUnlearner).

Key facts. Cited by 30 papers in the map (data through 14 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 42 of its 60 references that are in the map. First posted on arXiv 16 Feb 2025.

Builds on

42 of 60 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024Protecting Privacy in Multimodal Large Language Models with MLLMU-Bench29 Oct
  2. 2024Cross-Modal Safety Alignment: Is textual unlearning all you need?27 May
  3. 2024MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance5 Jan

Other references · 39

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  3. 2022BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generationcited by 4,322
  4. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  5. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  6. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  7. 2021Align before Fuse: Vision and Language Representation Learning with Momentum Distillationcited by 1,619
  8. 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,121
  9. 2019Machine Unlearningcited by 1,061
  10. 2024DeepSeek-VL: Towards Real-World Vision-Language Understandingcited by 733
  11. 2024TOFU: A Task of Fictitious Unlearning for LLMscited by 533
  12. 2024The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearningcited by 509
27 more
  1. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearningcited by 504
  2. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 386
  3. 2023Who's Harry Potter? Approximate Unlearning in LLMscited by 381
  4. 2023PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embeddingcited by 328
  5. 2021Unrolling SGD: Understanding Factors Influencing Machine Unlearningcited by 279
  6. 2023In-Context Unlearning: Language Models as Few Shot Unlearnerscited by 224
  7. 2024Towards Safer Large Language Models through Machine Unlearningcited by 139
  8. 2020Variational Bayesian Unlearningcited by 117
  9. 2023KGA: A General Machine Unlearning Framework Based on Knowledge Gap Alignmentcited by 99
  10. 2024Guardrail Baselines for Unlearning in LLMscited by 96
  11. 2023UrbanCLIP: Learning Text-enhanced Urban Region Profiling with Contrastive Language-Image Pretraining from the Webcited by 78
  12. 2024A Survey of Mathematical Reasoning in the Era of Multimodal Large Language Model: Benchmark, Method & Challengescited by 55
  13. 2024Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Surveycited by 52
  14. 2024Unified Gradient-Based Machine Unlearning with Remain Geometry Enhancementcited by 45
  15. 2024Digital Forgetting in Large Language Models: A Survey of Unlearning Methodscited by 40
  16. 2024Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Conceptscited by 40
  17. 2024CLEAR: Character Unlearning in Textual and Visual Modalitiescited by 39
  18. 2024ErrorRadar: Benchmarking Complex Mathematical Reasoning of Multimodal Large Language Models Via Error Detectioncited by 37
  19. 2024Revisiting Who's Harry Potter: Towards Targeted Unlearning from a Causal Intervention Perspectivecited by 36
  20. 2024Deep Learning for Cross-Domain Data Fusion in Urban Computing: Taxonomy, Advances, and Outlookcited by 33
  21. 2025Position: Multimodal Large Language Models Can Significantly Advance Scientific Reasoningcited by 33
  22. 2024Digger: Detecting Copyright Content Mis-usage in Large Language Model Trainingcited by 25
  23. 2024Multimodal Unlearnable Examples: Protecting Data against Multimodal Contrastive Learningcited by 21
  24. 2024CLIPErase: Efficient Unlearning of Visual-Textual Associations in CLIPcited by 19
  25. 2024Zero-Shot Class Unlearning in CLIP with Synthetic Samplescited by 17
  26. 2024From Efficient Multimodal Models to World Models: A Surveycited by 13
  27. 2024Learn while Unlearn: An Iterative Unlearning Framework for Generative Language Modelscited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 16; 2026: 14.

30 papers.

Sort:

By area:

  1. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deploymentcited by 116
  2. 2025LLM Agents for Education: Advances and Applicationscited by 59
  3. 2025MathAgent: Leveraging a Mixture-of-Math-Agent Framework for Real-World Multimodal Mathematical Error Detectioncited by 24
  4. 2025Are Multimodal Large Language Models Ready for Omnidirectional Spatial Reasoning?cited by 23
  5. 2025DocPruner: A Storage-Efficient Framework for Multi-Vector Visual Document Retrieval via Adaptive Patch-Level Embedding Pruningcited by 20
  6. 2025GUARD: Generation-time LLM Unlearning via Adaptive Restriction and Detectioncited by 17
  7. 2025Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-Tuning and Can Be Mitigated by Machine Unlearningcited by 16
  8. 2025Towards Omnidirectional Reasoning with 360-R1: A Dataset, Benchmark, and GRPO-based Methodcited by 13
  9. 2025HyperG: Hypergraph-Enhanced LLMs for Structured Knowledgecited by 11
  10. 2025MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steeringcited by 10
  11. 2026Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Frameworkcited by 10
  12. 2025OFFSIDE: Benchmarking Unlearning Misinformation in Multimodal Large Language Modelscited by 8
18 more
  1. 2025CAFES: A Collaborative Multi-Agent Framework for Multi-Granular Multimodal Essay Scoringcited by 7
  2. 2025SUA: Stealthy Multimodal Large Language Model Unlearning Attackcited by 4
  3. 2025Towards Reasoning-Preserving Unlearning in Multimodal Large Language Modelscited by 4
  4. 2026Visual-Guided Key-Token Regularization for Multimodal Large Language Model Unlearningcited by 3
  5. 2026Null Space Constrained Contrastive Visual Forgetting for MLLM Unlearningcited by 3
  6. 2026Position: The Term "Machine Unlearning" Is Overused in LLMscited by 3
  7. 2026Mitigating Memorization Where It Happenscited by 2
  8. 2025Robust MLLM Unlearning via Visual Knowledge Distillationcited by 1
  9. 2026ICU-Bench:Benchmarking Continual Unlearning in Multimodal Large Language Modelscited by 1
  10. 2026PPU-Bench:Real World Benchmark for Personalized Partial Unlearning in Vision Language Modelscited by 1
  11. 2026MLUBench: A Benchmark for Lifelong Unlearning Evaluation in MLLMscited by 1
  12. 2025Hierarchical Dual-Strategy Unlearning for Biomedical and Healthcare Intelligence Using Imperfect and Privacy-Sensitive Medical Datacited by 0
  13. 2026SPACE: Source-free Proxy Anchor Concept Erasure for MLLMscited by 0
  14. 2026POPS: Recovering Unlearned Multi-Modality Knowledge in MLLMs with Prompt-Optimized Parameter Shakingcited by 0
  15. 2026Stochastic Meta-Unlearning: Bridging Language Backbone and Multimodal Unlearningcited by 0
  16. 2026Partition-Aware Unlearning for Removing Spurious Correlations in Large Vision-Language Modelscited by 0
  17. 2026SIEVE: Selective attention-value Suppression for Vision-Language Models Unlearningcited by 0
  18. 2026CUSE: Fine-Grained Multimodal Unlearning with Counterfactual Guidance and On-Policy Self-Distillationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025PEBench: A Fictitious Dataset to Benchmark Machine Unlearning for Multimodal Large Language Models
  2. 2026Visual-Noise Guided In-Context Distillation for Multimodal Large Language Model Unlearning
  3. 2026Exploring and Bridging Knowledge Holes in Unlearned Multimodal Large Language Models
  4. 2024Single Image Unlearning: Efficient Machine Unlearning in Multimodal Large Language Models
  5. 2026Open-MMUnlearning: Unifying Methods and Evaluation for MLLM Unlearning
  6. 2026Knowledge Vector Weakening: Efficient Training-free Unlearning for Large Vision-Language Models
  7. 2025AUVIC: Adversarial Unlearning of Visual Concepts for Multi-modal Large Language Models
  8. 2025SafeEraser: Enhancing Safety in Multimodal Large Language Models through Multimodal Machine Unlearning