AI Paper Atlas

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

TRCE: Towards Reliable Malicious Concept Erasure in Text-to-Image Diffusion Models

Ruidong Chen, Honglin Guo, Lanjun Wang, Chenyu Zhang, Weizhi Nie, An-An Liu

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Concept erasure in diffusion models · 13 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 second-stage denoising trajectory steering to textual erasure (ablation)TRCE · sexual concept, I2PASR: 5.05% -> 1.29%Improved
add second-stage steering to textual erasure (ablation)TRCE · sexual concept, MMA-DiffusionASR: 7.80% -> 1.40%Improved
add first-stage textual erasure to trajectory steering (ablation)TRCE · sexual concept, I2PASR: 13.86% -> 1.29%Improved
add second stage at high knowledge preservation rate eta=0.02 (ablation)TRCE · adversarial promptsAdv ASR: 13.52% -> 2.36%Improved
optimize [EoT] embedding instead of [KEY] only (ablation)TRCE · I2P sexualASR: 22.8% -> 5.05%Improved
also optimize [KEY] with [EoT] (ablation)TRCE · I2P sexualASR: 5.05% -> 6.34%Hurt
add guidance enhancement in second stage (ablation)TRCE · I2P sexualASR: 4.65% -> 1.29%Improved
add L_preserve in second stage (ablation)TRCE · I2P sexualFID_gen: 14.64 -> 12.08Improved
add contrastive L_erase instead of safe alignment only (ablation)TRCE · I2P sexualASR: 1.93% -> 1.29%Improved
raise knowledge preservation rate eta from 0.002 to 0.02 (stage one only) (ablation)TRCE · I2P sexualASR: 3.54% -> 7.20%Hurt
use TRCE vs MACE for multi-concept erasureTRCE · I2P 7 categoriesoverall inappropriate rate: 5.6% -> 2.0%Improved
use TRCE(T+V) vs AdvUnlearn on UnlearnDiff attackTRCE · sexual conceptASR: 3.52% -> 0.70%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