Contrastive Pre-training for Discourse-Aware Conversation Models
June 15, 2021
roland-laboulaye.jpg

Thursday, July 8th at 11:00am, 3350 TMCB

Advisor: Kevin Seppi

MS Thesis Defense for Roland Laboulaye

Abstract:

Understanding long conversations requires recognizing a discourse flow unique to conversation. Recent advances in unsupervised representation learning of text have been attained primarily through language modeling, which models discourse only implicitly and within a small window. These representations are in turn evaluated chiefly on sentence pair or paragraph-question pair benchmarks, which measure only local discourse coherence.

In order to improve performance on discoursereliant, long conversation tasks, we propose Turn-of-Phrase pre-training, an objective designed to encode long conversation discourse flow. We leverage tree-structured Reddit conversations in English to, relative to a chosen conversation path through the tree, select paths of varying degrees of relatedness. The final utterance of the chosen path is appended to the related paths and the model learns to identify the most coherent conversation path.

We demonstrate that our pre-training objective encodes conversational discourse aware- ness by improving performance on a dialogue act classification task. We then demonstrate the value of transferring discourse awareness with a comprehensive array of conversationlevel classification tasks evaluating persuasion, conflict, and deception.