AI Paper Atlas

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

Where Concept Erasure Should Occur: Concept–Layer Alignment in Text-to-Video Diffusion Models

Yiwei Xie, Ping Liu, Zheng Zhang

ICML 2026 · Image generation › Text-to-image diffusion · approach: Multi-concept composition · 0 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 separability contrastive loss Lcon to layer search (ablation)CLEAR · Parachute, Wan2.2-5Bgenerative rate: 27.0% -> 18.4%Improved
add separability contrastive loss Lcon to layer search (ablation)CLEAR · Springer dog, Wan2.2-5Bgenerative rate: 14.0% -> 4.8%Improved
add separability contrastive loss Lcon to layer search (ablation)CLEAR · Nudity, Wan2.2-5Bgenerative rate: 9.4% -> 10.9%Hurt
intervene at top-1 only instead of top-1 and top-2 layers (ablation)CLEAR · Nudity, Wan2.2-5Bgenerative rate: 30.1% -> 11.1%Improved
increase intervention strength gamma from 8 to 12 (ablation)CLEAR · Nudity, Wan2.2-5Bgenerative rate: 28.4% -> 3.2%Improved
increase intervention strength gamma from 10 to 12 (ablation)CLEAR · Nudity, Wan2.2-5Bconsistency: 0.1754 -> 0.0933Hurt
use sparse autoencoder features instead of dense steering vector (ablation)CLEAR · Nudity, Wan2.2-5Bgenerative rate: 24.5% -> 11.1%Improved
learned differentiable layer search instead of brute-force search (ablation)CLEAR · Springer dog, Parachute, Nuditysearch cost: 24x -> 1xImproved
use CLEAR instead of fine-tuning (T2VUnlearning) on small modelCLEAR · Object erasure, CogVideoX-2Bimaging quality: 0.3907 -> 0.4683Improved
use CLEAR instead of T2VUnlearningCLEAR · Object erasure, Wan2.2-5Bgenerative rate: 24.5% -> 12.8%Improved
use CLEAR instead of VideoEraserCLEAR · Celebrity erasure, CogVideoX-2Bidentity similarity: 0.1140 -> 0.1802Hurt
apply CLEAR to evasive Ring-A-Bell promptsCLEAR · Ring-A-Bell, Wan2.2-5Bgenerative rate: 69.6% -> 25.6%Improved

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