mlsanityData to 14 Oct 2026

AIM: Anchor Identity Features, Then Match for Multimodal Large Language Model Unlearning

arXiv 2608.28312 ↗First posted 28 Aug 2026cs.CV cs.CLcited by 1

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

also in 1 more
Abstract (arXiv)

Multimodal large language models (MLLMs) can memorize identity-specific facts about people in their fine-tuning data, creating privacy risks when a person requests deletion. Existing MLLM unlearning methods often assume access to retain images or ground-truth answers during deletion, which is unrealistic in many practical scenarios. We study identity unlearning when retain images are unavailable at deletion time. Our analysis shows that identity and visual-perception questions occupy distinct regions in fine-tuned hidden states and are organized differently: identity questions cluster by person, whereas perception questions cluster by question type. This suggests that identity knowledge can be suppressed without erasing general visual perception. Building on this observation, we propose AIM, a two-stage method that anchors an identity-forgetting target with a universal visual prompt and then matches the vision encoder to that target under a Fisher-based constraint. Extensive experiments show that AIM achieves competitive identity forgetting while preserving non-deleted identities, prior knowledge, and visual perception on the same images.

Key facts. Cited by 1 paper 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 27 of its 40 references that are in the map. First posted on arXiv 28 Aug 2026.

Builds on

27 of 40 references.

Closest ancestors · 6

Same direction, newest first

  1. 2026Before Forgetting, Learn to Remember: Revisiting Foundational Learning Failures in LVLM Unlearning Benchmarks5 May
  2. 2025Towards Benign Memory Forgetting for Selective Multimodal Large Language Model Unlearning25 Nov
  3. 2025SineProject: Machine Unlearning for Stable Vision Language Alignment23 Nov
  4. 2025Cross-Modal Unlearning via Influential Neuron Path Editing in Multimodal Large Language Models10 Nov
  5. 2025SAUCE: Selective Concept Unlearning in Vision-Language Models with Sparse Autoencoders16 Mar
  6. 2024Protecting Privacy in Multimodal Large Language Models with MLLMU-Bench29 Oct

Other references · 21

  1. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  2. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  3. 2016Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answeringcited by 2,516
  4. 2020Extracting Training Data from Large Language Modelscited by 2,189
  5. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,105
  6. 2019Machine Unlearningcited by 1,061
  7. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 864
  8. 2015Optimizing Neural Networks with Kronecker-factored Approximate Curvaturecited by 796
  9. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 618
  10. 2023Dissecting Recall of Factual Associations in Auto-Regressive Language Modelscited by 528
  11. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearningcited by 504
  12. 2019PyHessian: Neural Networks Through the Lens of the Hessiancited by 292
9 more
  1. 2021Unrolling SGD: Understanding Factors Influencing Machine Unlearningcited by 279
  2. 2013Revisiting Natural Gradient for Deep Networkscited by 159
  3. 2020Variational Bayesian Unlearningcited by 117
  4. 2025Overcoming catastrophic forgetting in neural networkscited by 55
  5. 2024On Large Language Model Continual Unlearningcited by 52
  6. 2024Remaining-data-free Machine Unlearning by Suppressing Sample Contributioncited by 13
  7. 2024Forget Vectors at Play: Universal Input Perturbations Driving Machine Unlearning in Image Classificationcited by 6
  8. 2026Gauss-Newton Unlearning for the LLM Eracited by 4
  9. 2026Visual-Guided Key-Token Regularization for Multimodal Large Language Model Unlearningcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Where Identity Lives: Localized, Retain-Free Identity Unlearning in Multimodal Large Language Modelscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026Toward Fine-Grained Forgetting:Attribute Unlearning for Multimodal Large Language Models
  2. 2024Single Image Unlearning: Efficient Machine Unlearning in Multimodal Large Language Models
  3. 2026SIEVE: Selective attention-value Suppression for Vision-Language Models Unlearning
  4. 2025Rethinking Post-Unlearning Behavior of Large Vision-Language Models
  5. 2024Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset
  6. 2026ICU-Bench:Benchmarking Continual Unlearning in Multimodal Large Language Models
  7. 2026PPU-Bench:Real World Benchmark for Personalized Partial Unlearning in Vision Language Models
  8. 2025AUVIC: Adversarial Unlearning of Visual Concepts for Multi-modal Large Language Models