🛰️ Daily AI Frontier
‹ back to 2026-09-01

Where Identity Lives: Localized, Retain-Free Identity Unlearning in Multimodal Large Language Models

Research Multimodal & Generative

Ranking

Overall 81
Content 100
Popularity 37

Observed public metrics from 1 member.

Merged summary

TL;DR - PAVA is a retain-free method for removing an individual’s identity information from multimodal large language models while preserving visual-language capabilities. It matters because it avoids reconstructing privacy-sensitive retain sets after deployment.

  • Causal tracing, weight transplant, and Fisher overlap localize identity information primarily to early-to-mid decoder MLP layers.
  • PAVA restricts updates to those layers to reduce disruption to shared visual-language processing.
  • A visual-attribute anchor distills the model’s pre-unlearning, image-grounded answers using only forget-set images.
  • On MLLMU-Bench and ReMem, PAVA achieves the strongest forget-retain trade-off among forget-set-only methods and remains competitive with retain-based baselines.

Sources (1)

Where Identity Lives: Localized, Retain-Free Identity Unlearning in Multimodal Large Language Models

arXiv cs.CL Kangwook Ko, Jaehyuk Jang, Wonjun Lee, Hee-Seon Kim, Changick Kim 2026-08-31 arXiv:2608.30649
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-24 14:30:05.087051 UTC

TL;DR - PAVA is a retain-free method for removing an individual’s identity information from multimodal large language models while preserving visual-language capabilities. It matters because it avoids reconstructing privacy-sensitive retain sets after deployment.

  • Causal tracing, weight transplant, and Fisher overlap localize identity information primarily to early-to-mid decoder MLP layers.
  • PAVA restricts updates to those layers to reduce disruption to shared visual-language processing.
  • A visual-attribute anchor distills the model’s pre-unlearning, image-grounded answers using only forget-set images.
  • On MLLMU-Bench and ReMem, PAVA achieves the strongest forget-retain trade-off among forget-set-only methods and remains competitive with retain-based baselines.
item →