AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Robust Domain Adaptation for Machine Reading Comprehension

Liang Jiang, Zhenyu Huang, Jia Liu, Zujie Wen, Xi Peng

AAAI 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 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
Construct pseudo QA pairs from documents and dialogue, with answer filtering, question selection, and QS trainingBERT MRC · d-QuAC ALLF1: 19.33 -> 24.46Improved
Construct pseudo QA pairs from documents and dialogue, with answer filtering, question selection, and QS trainingBERT MRC · d-CoQA ALLF1: 43.27 -> 49.35Improved
Construct pseudo QA pairs from documents and dialogue, with answer filtering, question selection, and QS trainingBERT MRC · AlipayF1: 44.28 -> 55.67Improved
Add reward-baseline term r_b to QS training (ablation)RMRC · AlipayF1: 54.13 -> 55.67Improved
Add answer filtering (ablation)RMRC · AlipayF1: 35.05 -> 55.67Improved
Add question fusing (ablation)RMRC · AlipayF1: 53.71 -> 55.67Improved
Add QS training (ablation)RMRC · AlipayF1: 51.72 -> 55.67Improved
Use reinforced self-training instead of confidence-based self-training (ablation)RMRC · AlipayF1: 52.34 -> 55.67Improved
Use F1 rather than cross-entropy as the QS reward (ablation)RMRC · AlipayEM: 7.71 -> 9.28Improved
Use F1 rather than cross-entropy as the QS reward (ablation)RMRC · AlipayF1: 55.52 -> 55.67Improved

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