Using dialogue act prediction as an auxiliary task during training lets a dialogue coherence model rank dialogues without explicit dialogue act labels at evaluation.
In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, 31 October – 4 November 2018, pages 4328–4339
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
-
Dialogue Coherence Assessment Without Explicit Dialogue Act Labels
Using dialogue act prediction as an auxiliary task during training lets a dialogue coherence model rank dialogues without explicit dialogue act labels at evaluation.