AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Hierarchical Attention Network with Pairwise Loss for Chinese Zero Pronoun Resolution

Peiqin Lin, Meng Yang

AAAI 2020 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Robust fine-tuning of zero-shot models · 6 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 self-attention layer (ablation)HAN-PL · OntoNotes 5.0 Chinese test, OverallF-score (%): Table 3: 58.6 vs 60.2Improved
add npc-to-zp attention (ablation)HAN-PL · OntoNotes 5.0 Chinese test, OverallF-score (%): Table 3: 59.4 vs 60.2Improved
add zp-to-npc attention (ablation)HAN-PL · OntoNotes 5.0 Chinese test, OverallF-score (%): Table 3: 58.2 vs 60.2; zp-to-npc more importantImproved
use pairwise-margin loss instead of softmax cross entropy (ablation)HAN-PL · OntoNotes 5.0 Chinese test, OverallF-score (%): performance falls sharply by 3.7%Improved
add similarity constraint among correct antecedents (ablation)HAN-PL · OntoNotes 5.0 Chinese test, OverallF-score (%): Table 4: 58.6 vs 60.2Improved
HAN-PL vs best baseline Yin et al. (2018b), overallHAN-PL · OntoNotes 5.0 Chinese test, OverallF-score (%): outperforms the best baseline by 2.9%Improved
HAN-PL vs best baseline on source MZHAN-PL · OntoNotes 5.0 Chinese test, MZF-score (%): 52.5 -> 52.6No clear effect
HAN-PL vs best baseline on source TCHAN-PL · OntoNotes 5.0 Chinese test, TCF-score (%): 53.2 -> 54.7Improved
HAN-PL vs Yin et al. (2018b)HAN-PL · OntoNotes 5.0 Chinese test, BNF-score (%): 5.6% (source BN)Improved
larger candidate set size nc (harder resolution); HAN-PL gains over AttentionZP growHAN-PL · OntoNotes 5.0 candidate-set sizeF-score (%)Improved

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