AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

MultiDelete for Multimodal Machine Unlearning

Jiali Cheng, Hadi Amiri

ECCV 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Unified any-modality language models · 29 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add modality decoupling loss (ablation)MultiDelete · NLVR2 (ALBEF)deletion set accuracy (Df): 50.3 -> 23.5Improved
add unimodal knowledge retention loss (ablation)MultiDelete · NLVR2 (ALBEF)test accuracy (DTest): 79.2 -> 80.8Improved
add multimodal knowledge retention loss (ablation)MultiDelete · NLVR2 (ALBEF)test accuracy (DTest): 77.1 -> 80.8Improved
add unimodal knowledge retention loss (ablation)MultiDelete · Unimodal image classifier accuracyaccuracy: 77.9 -> 83.6Improved
multimodal unlearning with MultiDelete instead of FineTuneALBEF · NLVR2 deletion setaccuracy (Df): 49.8 -> 23.5Improved
weight scrubbing (DtD, L-codec) on all parametersmultimodal models · Flickr30K-IR (ALBEF)test recall (DTest): 97.8 (Retrain) -> 10.3 (DtD)Hurt
update only the fusion module instead of all parameters (ablation)MultiDelete · average over tasks (ALBEF)test performance: 84.2 -> 83.7No clear effect
update only the fusion module for scrubbing methods (ablation)DtD · average over tasks (ALBEF)test performance: 33.0 -> 39.2Improved
update only the fusion module for modality-agnostic approaches (ablation)FineTune, NegGrad, UL · average over tasks (ALBEF)test performance: marginal gap of less than 0.6 absolute pointsNo clear effect
use MultiDelete instead of L-codec for membership inference protectionALBEF · average MI ratioprior-to-post MI ratio: 1.19 -> 1.27Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy