AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

LEACE: Perfect linear concept erasure in closed form

Nora Belrose, David Schneider-Joseph, Shauli Ravfogel, Ryan Cotterell, Edward Raff, Stella Biderman

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Prototype and concept dictionary 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
use LEACE instead of INLP/RLACE for gender erasureBERT [CLS] representation · Biographies gender erasuregender prediction accuracy vs edit sizeImproved
apply LEACE to BERT representationsprofession classifier · BiographiesTPR gap RMS: 0.198 -> 0.084Improved
apply LEACE to BERT representationsprofession classifier · Biographiesprofession accuracy: 79.3 -> 77.3Hurt
apply LEACE to BERT representationsprofession classifier · Biographiescorrelation of TPR gap with % women: 0.867 -> 0.392Improved
refit profession classifier on projected representations (ablation)LEACE · Biographiesprofession accuracy and TPR gap: 77.3 -> 78.1 acc; 0.084 -> 0.158 TPR gapMixed
use LEACE instead of SAL for concept scrubbing (ablation)LLaMA/Pythia · POS scrubbingperplexity increase (lower is less damage): LLaMA 7B 3.24 -> 1.73Improved
erase POS with LEACE vs random same-rank subspace (ablation)LLaMA 7B · POS scrubbingperplexity (bits per byte): 0.69 -> 1.73Hurt
LEACE instead of INLP for erasing POS from BERT layerBERT amnesic probing · POS erasurerank reduction: 360 -> 17Improved

Later papers that compared with this method

4 later accepted papers report LEACE as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

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