Topic-Aware Contrastive Learning for Abstractive Dialogue Summarization

Unlike well-structured text, such as news reports and encyclopedia articles, dialogue content often comes from two or more interlocutors, exchanging information with each other. In such a scenario, the topic of a conversation can vary upon progression and the key information for a certain topic is often scattered across multiple utterances of different speakers, which poses challenges to abstractly summarize dialogues. To capture the various topic information of a conversation and outline salient facts for the captured topics, this work proposes two topic-aware contrastive learning objectives, namely coherence detection and sub-summary generation objectives, which are expected to implicitly model the topic change and handle information scattering challenges for the dialogue summarization task. The proposed contrastive objectives are framed as auxiliary tasks for the primary dialogue summarization task, united via an alternative parameter updating strategy. Extensive experiments on benchmark datasets demonstrate that the proposed simple method significantly outperforms strong baselines and achieves new state-of-the-art performance. The code and trained models are publicly available via \href{}{}.

PDF Abstract Findings (EMNLP) 2021 PDF Findings (EMNLP) 2021 Abstract


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Text Summarization SAMSum Corpus ConDigSum ROUGE-1 54.3 # 3
ROUGE-2 29.3 # 3
ROUGE-L 45.2 # 5
BertScoreF1 54 # 2
Text Summarization SAMSum Corpus BART-Large ROUGE-1 52.6 # 6
ROUGE-2 27 # 6
ROUGE-L 42.1 # 6
BertScoreF1 52.1 # 3