Dial2vec: Self-Guided Contrastive Learning of Unsupervised Dialogue Embeddings

TL;DR

This paper introduces dial2vec, a self-guided contrastive learning method for unsupervised dialogue embeddings that effectively captures conversational interactions, significantly improving performance on multiple dialogue understanding tasks.

Contribution

The paper proposes dial2vec, a novel contrastive learning approach that models conversational interactions for unsupervised dialogue embedding learning, outperforming existing methods.

Findings

Dial2vec achieves significant improvements in dialogue understanding tasks.

It produces more informative and discriminative embeddings.

The approach effectively captures conversational interaction patterns.

Abstract

In this paper, we introduce the task of learning unsupervised dialogue embeddings. Trivial approaches such as combining pre-trained word or sentence embeddings and encoding through pre-trained language models (PLMs) have been shown to be feasible for this task. However, these approaches typically ignore the conversational interactions between interlocutors, resulting in poor performance. To address this issue, we proposed a self-guided contrastive learning approach named dial2vec. Dial2vec considers a dialogue as an information exchange process. It captures the conversational interaction patterns between interlocutors and leverages them to guide the learning of the embeddings corresponding to each interlocutor. The dialogue embedding is obtained by an aggregation of the embeddings from all interlocutors. To verify our approach, we establish a comprehensive benchmark consisting of six widely-used dialogue datasets. We consider three evaluation tasks: domain categorization, semantic relatedness, and dialogue retrieval. Dial2vec achieves on average 8.7, 9.0, and 13.8 points absolute improvements in terms of purity, Spearman's correlation, and mean average precision (MAP) over the strongest baseline on the three tasks respectively. Further analysis shows that dial2vec obtains informative and discriminative embeddings for both interlocutors under the guidance of the conversational interactions and achieves the best performance when aggregating them through the interlocutor-level pooling strategy. All codes and data are publicly available at https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/dial2vec.