AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

VL-Eraser: Vacuum Distillation for Machine Unlearning in Vision-Language Models

Yili Wang, Lu Dai, Tairan Huang, Yijie Xu, Hui Xiong

CVPR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Test-time adaptation of VLMs

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 vacuum-space projection to distillation (ablation)VL-Eraser · MLLMU-Bench Visual-QA retain setclassification accuracy (%): 36.6 -> 43.4Improved
add distillation loss to cross-entropy (ablation)VL-Eraser · MLLMU-Bench Visual-QA forget setclassification accuracy (%): 38.4 -> 26.2 (lower is better)Improved
add cross-entropy loss to distillation (ablation)VL-Eraser · MLLMU-Bench Visual-QA forget setclassification accuracy (%): 32.6 -> 26.2 (lower is better)Improved
distill-then-delete instead of reverse training (vs GA/NPO)LLaVA-1.5-7B · MLLMU-Bench Textual-QA forget setclassification accuracy (%): 43.6 (NPO) -> 25.0Improved
VL-Eraser vs MMUnlearnerQwen2-VL-7B · MLLMU-Bench Visual-QA forget setclassification accuracy (%): 40.2 -> 33.6Improved
VL-Eraser utility on real celebrity setLLaVA-1.5-7B · MLLMU-Bench Visual-QA real setclassification accuracy (%): 47.2 -> 45.6Mixed

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