AI Paper Atlas

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

Trustworthy Classification for Complex Social Surveys: A Memory-Enhanced Hierarchical Framework with Calibrated Uncertainty

Zeqiang Wang, Rebecca Oldroyd, Yuqi Wang, Jiageng Wu, Jie Yang, Wei Wang, Nishanth R. Sastry, Jon Johnson, Suparna De

AAAI 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Group-robust classification

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 HOCM to BERTBERT · Social survey in-dist testAccuracy (%): 85.3 -> 84.9Hurt
add HOCM to BERT, evaluate on confident subsetBERT · Social survey in-dist testCorrected accuracy (%): 84.9 -> 92.1Improved
add HOCM to BERTBERT · Social survey open-set detectionMVR (%): 68.0 -> 78.9Improved
HOCM vs dedicated OSR methods (LOS)BERT · Social survey open-set detectionMVR (%): 78.8 -> 78.9No clear effect
add HOCM to CharCNN+BiLSTMCharCNN+BiLSTM · Social survey in-dist testAccuracy (%): 77.1 -> 80.7Improved
add memory-enhanced contrastive learning (ablation)BERT + hierarchical UC · Social survey ablationCorrected accuracy (%): 89.0 -> 92.1Improved
add memory-enhanced contrastive learning (ablation)BERT + hierarchical UC · Social survey ablationMVR (%): 74.5 -> 78.9Improved
add quality filter in memory management (ablation)HOCM · Social survey ablationMVR (%): 78.0 -> 78.9Improved
add reliability-guided sampling (ablation)HOCM · Social survey ablationMVR (%): 78.2 -> 78.9Improved
use hierarchical calibration instead of flat OpenMax (ablation)HOCM · Social survey ablationMVR (%): 76.5 -> 78.9Improved
add hierarchical uncertainty calibration (ablation)BERT + Memory CL · Social survey ablationMVR (%): 72.5 -> 78.9Improved
add memory CL on closed-set accuracy (ablation)BERT + hierarchical UC · Social survey ablationAccuracy (%): 85.1 -> 84.9Hurt

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