AI Paper Atlas

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

ForceForget: Reinforcement Concept Removal for Enhancing Safety in Text-to-Image Models

Dong Han, Yong Li

ICML 2026 · Image generation › Text-to-image diffusion · approach: Concept erasure in diffusion models

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
raise unsafe-score weight in Safety Evaluator from -1 to -2 (ablation)ForceForget · I2P (sexual)NRR: 82.92 -> 97.96Improved
use balanced CER weights (1,1) instead of (2,1) (ablation)ForceForget · I2P (sexual)NRR: 81.95 -> 97.96Improved
use 4 safe-adapter tokens instead of 2 (ablation)ForceForget · I2P (sexual)NRR: 87.43 -> 97.96Improved
use 4 safe-adapter tokens instead of 8 (ablation)ForceForget · I2P (sexual)NRR: 83.67 -> 97.96Improved
use 4 safe-adapter tokens instead of 16 (ablation)ForceForget · COCO-30kCLIP score: 27.924 -> 30.786Improved
apply safe adapter to end tokens rather than beginning tokens (ablation)ForceForget · COCO-30kCLIP score: 18.850 -> 30.786Improved
apply adapter to 32 end tokens rather than 32 beginning tokens (ablation)ForceForget · I2P (sexual)NRR: 92.80 -> 83.35Hurt
add alignment reward to safety reward (ablation)ForceForget · COCO-30kCLIP score: 21.1603 -> 30.53Improved
add safety reward to alignment reward (ablation)ForceForget · I2P (sexual)NRR: 93.89 -> 97.96Improved
add safe adapter to DDPO fine-tuning (ablation)ForceForget · I2P (sexual)reward mean / erasingImproved
combine BLIP caption and target condition in alignment reward (ablation)ForceForget · I2P (sexual)erase-preserve tradeoffImproved
transfer LoRA trained on SD v1.4 to SD v1.5ForceForget · I2P (sexual)NRR: 78.3 -> 99.2Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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