AI Paper Atlas

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

CAWA: An Attention-Network for Credit Attribution

Saurav Manchanda, George Karypis

AAAI 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 2 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 CAWA instead of MLTM for sentence attributionCredit attribution · Ohsumed, TMC2007, Patents, DeliciousPPPA: average performance gain of 6.2%Improved
Use CAWA instead of SEG-REFINE for sentence attributionCredit attribution · Ohsumed, TMC2007, Patents, DeliciousPPPA: average performance gain of 9.8%Improved
Use CAWA instead of MLTMMultilabel classification · Five datasetsF1: average performance gain of 4.1%Improved
Use CAWA instead of SEG-REFINEMultilabel classification · Five datasetsF1: average performance gain of 1.6%Improved
Add attention to the DNN-A architecture (ablation)DNN-A · All five datasetsSOV and PPPAImproved
Add average pooling before softmax (ablation)CAWA · All five datasets; varying βSOVImproved
Increase β too far, making attention outweigh document probabilities (ablation)CAWA · Credit attribution; large βSOVHurt
Increase α, reducing the relative contribution of attention loss (ablation)CAWA · All five datasetsAUCImproved
Add attention loss that penalizes attention on absent classes (ablation)CAWA · Multilabel classificationAUCHurt
Model each sentence as a distribution over classes rather than a single classCAWA · Ohsumed, Patents, Delicious similar-class subsetsF1: 0.64 -> 0.68; 0.48 -> 0.53; 0.26 -> 0.39; 0.33 -> 0.41; 0.47 -> 0.50; 0.58 -> 0.62Improved

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