A self-attention reading comprehension model that injects discourse, semantic role, and coreference annotations into dedicated attention heads improves NarrativeQA question answering by up to 3.4 Rouge-L over the QANet baseline.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Discourse-Aware Semantic Self-Attention for Narrative Reading Comprehension
A self-attention reading comprehension model that injects discourse, semantic role, and coreference annotations into dedicated attention heads improves NarrativeQA question answering by up to 3.4 Rouge-L over the QANet baseline.