AI Paper Atlas

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

Learning Concept Credible Models for Mitigating Shortcuts

Jiaxuan Wang, Sarah Jabbour, Maggie Makar, Michael Sjoding, Jenna Wiens

NeurIPS 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Prototype and concept dictionary models · 1 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 EYE regularization (lambda>0) to STD(C,X) (ablation)CCM EYE vs STD(C,X) · CUB clean testaccuracy: 69.7 -> 75.2Improved
condition on known concepts C in addition to XSTD(X) -> STD(C,X) · CUB clean testaccuracy: 55.8 -> 69.7Improved
use CCM RES instead of standard training on XSTD(X) · CUB clean testaccuracy: 55.8 -> 76.0Improved
use CCM EYE instead of standard training on XSTD(X) · CUB biased testaccuracy: 75.7 -> 76.0No clear effect
use CBM (concepts only) instead of CCMCBM · CUB biased testaccuracy: 76.0 -> 71.6Hurt
learn C on biased data (vary T), violating A1CCM · CUB clean testaccuracyMixed
replace dimensions of C with noise, violating A2CCM · CUB clean testaccuracyMixed
CCM EYE on MIMIC with sex shortcut vs baselinesCCM EYE · MIMIC-CXR edemaaccuracy under shifted correlationImproved
reverse test correlation relative to trainingall methods · MIMIC-CXR edemaaccuracyHurt

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