AI Paper Atlas

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

Prototype-Guided Concept Erasure in Diffusion Models

Yuze Cai, Jiahao Lu, Hongxiang Shi, Yichao Zhou, Hong Lu

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Concept erasure in diffusion models · 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
replace single-direction erasure with a bank of clustered concept prototypes as negative guidanceSD v1.4 · I2P (overall)inappropriate rate (%): 35.6% -> 5.2%Improved
apply prototype guidance on SDXL instead of SafreeSDXL · UnlearnDiff attackASR: 0.25 -> 0.13Improved
apply prototype guidance on SD 3.5 instead of SafreeSD 3.5 · P4D attackASR: 0.27 -> 0.09Improved
apply prototype guidance without adversarial-specific designSD v1.4 · Ring-a-Bell attackASR (%): 71.3% -> 6.7%Improved
use prototype guidance instead of TRCE on P4DSD v1.4 · P4D attackASR (%): 2.0% -> 14.5%Hurt
use prototype guidance instead of TRCE on FIDSD v1.4 · adversarial evaluationFID: 48.7 -> 45.1Improved
use prototype guidance instead of Safree on FIDSD v1.4 · adversarial evaluationFID: 36.3 -> 45.1Hurt
increase number of prototypes from 1 to 16 (ablation)prototype guidance · I2P sexualNudeNet rate and CLIP scoreMixed
increase number of prototypes beyond 16 (up to 128) (ablation)prototype guidance · I2P sexualCLIP score and NudeNetNo clear effect
use 16 prototypes for the shocking concept (ablation)prototype guidance · I2P shockingQ16 and CLIP scoreMixed
use prototype guidance for multi-style erasureSD v1.4 · Multi-style erasureAesthetic score: 5.84 -> 5.89Improved

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