AI Paper Atlas

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

Erasing More Than Intended? How Concept Erasure Degrades the Generation of Non-Target Concepts

Ibtihel Amara, Ahmed Imtiaz Humayun, Ivana Kajic, Zarana Parekh, Natalie Harris, Sarah Young, Chirag Nagpal, Najoung Kim, Junfeng He, Cristina Nader Vasconcelos, Deepak Ramachandran, Golnoosh Farnadi and 3 more

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
erase a concept with UCE (sensitivity vs original SD)UCE · EraseBench Visual Similarity (Object)CLIP sensitivity: 85.0 -> 82.7Hurt
erase a concept with Receler (sensitivity vs original SD)Receler · EraseBench Subset-SupersetCLIP sensitivity: 92.1 -> 36.7Hurt
erase a concept with AdvUnlearn (sensitivity vs original SD)AdvUnlearn · EraseBench Artistic Style SimilarityCLIP sensitivity: 69.0 -> 27.4Hurt
erase a concept with ESD (sensitivity vs original SD)ESD · EraseBench BinomialCLIP sensitivity: 88.5 -> 70.6Hurt
erase concepts with UCE, Gecko score on non-erased promptsUCE · EraseBench GeckoGecko alignment: 77.6 -> 74.3Hurt
erase concepts with MACE, Gecko score on non-erased promptsMACE · EraseBench GeckoGecko alignment: 77.6 -> 67.9Hurt
erase concepts with AdvUnlearn, Gecko score on non-erased promptsAdvUnlearn · EraseBench GeckoGecko alignment: 77.6 -> 68.6Hurt
erase concepts with MACE, object presence question accuracy on visually similar promptsMACE · EraseBench Gecko object presenceGecko object-presence accuracy: 89.0% -> 67.1%Hurt
erase concepts with UCE, object presence question accuracy on visually similar promptsUCE · EraseBench Gecko object presenceGecko object-presence accuracy: 89.0% -> 80.3%Hurt
erase 'cat' with UCE, RAHF artifact scoreUCE · EraseBench Erase CatRAHF artifacts: 87.71 -> 72.44Hurt
erase 'goldfish' with AdvUnlearn, RAHF aestheticsAdvUnlearn · EraseBench Erase GoldfishRAHF aesthetics: 78.21 -> 72.49Hurt
add related non-target concepts (tiger, panther) to the retain set (ablation)erasure with retain set · EraseBench erase catover-erasure and artifacts: partially alleviates over-erasure, artifacts persist, erased concept leaks backMixed

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