BERT fine-tuning substantially improves non-factoid passage re-ranking over prior baselines, with a 256-token input window performing best and chunking providing a workaround for longer passages.
Bruce Croft
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Study of BERT for Non-Factoid Question-Answering under Passage Length Constraints
BERT fine-tuning substantially improves non-factoid passage re-ranking over prior baselines, with a 256-token input window performing best and chunking providing a workaround for longer passages.