AI Paper Atlas

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

Disentangling Cognitive Diagnosis with Limited Exercise Labels

Xiangzhi Chen, Le Wu, Fei Liu, Lei Chen, Kun Zhang, Richang Hong, Meng Wang

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 5 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 margin loss in alignment module (ablation)DCD · Junyi 10% Q-matrixDOA: 0.7136 -> 0.7174Improved
add L2 loss in alignment module (ablation)DCD · Junyi 10% Q-matrixDOA: 0.6322 -> 0.7174Improved
add disentanglement module (ablation)DCD · Junyi 10% Q-matrixDOA: 0.7134 -> 0.7174Improved
add L2 loss in alignment module (ablation)DCD · Junyi 10% Q-matrixAUC: 0.8101 -> 0.8143Improved
add disentanglement module (ablation)DCD · Junyi 10% Q-matrixAUC: 0.8130 -> 0.8143Improved
add margin loss in alignment module (ablation)DCD · Junyi 10% Q-matrixAUC: 0.8140 -> 0.8143No clear effect
raise disentanglement weight beta from 0 to 2 (coarse levels)DCD · Junyi / NIPS2020EC few-labeledAUC and DOAMixed
disentangle only the finest concept level as beta risesDCD · Junyi / NIPS2020EC few-labeledDOANo clear effect
increase alignment loss weight alphaDCD · Junyi 10% Q-matrixAUCImproved
increase margin loss weight lambda1 from zero to nonzero, then excessiveDCD · Junyi 10% Q-matrixAUCMixed
increase d2 from 0 to 8DCD · Junyi 10% Q-matrixAUCMixed
DCD instead of KaNCD in 10% Q-matrix scenarioKaNCD · Matmat 10% Q-matrixDOA: 0.783 -> 0.796Improved

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