AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Sculpting Memory: Multi-Concept Forgetting in Diffusion Models via Dynamic Mask and Concept-Aware Optimization

Gen Li, Yang Xiao, Jie Ji, Kaiyuan Deng, Bo Hui, Linke Guo, Xiaolong Ma

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Concept erasure in diffusion models · 11 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 dynamic mask instead of fixed mask (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc: 0.175 -> 0.084Improved
add dynamic mask instead of fixed mask (ablation)Ours · Imagenette (unlearn 3 classes)Total Acc: 0.194 -> 0.093Improved
raise cross-attention sparsity from 20% to 50% (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc: 0.215 -> 0.084Improved
raise cross-attention sparsity from 50% to 80% (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc: 0.084 -> 0.110Hurt
restrict sparse updates to K and V matrices (50%) instead of cross-attention layers (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc: 0.084 -> 0.209Hurt
apply sparsity to whole UNet instead of cross-attention (50%) (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc / CLIP: 0.084 / 26.43 -> 0.001 / 19.91Mixed
mask update every 100 steps instead of 50 (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc: 0.120 -> 0.084Improved
mask update every 100 steps instead of 400 (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc: 0.125 -> 0.084Improved
mask update every 100 steps instead of 200 (ablation)Ours · Imagenette (unlearn 10 classes)Total Acc: 0.095 -> 0.084Improved
add regularization (distillation) loss with beta 0.25 (ablation)Ours · Imagenette (unlearn 10 classes)Total AccImproved
increase alignment loss weight alpha beyond 0.25 (ablation)Ours · Imagenette (unlearn 10 classes)Total AccImproved
use Ours instead of MACE on NSFW (I2P)Ours vs MACE · I2P NudeNet detection (all prompts)Total NSFW detections: 111 -> 47Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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