Instructive Dialogue Summarization with Query Aggregations
This work addresses the challenge of personalized dialogue summarization for users with specific interests, representing an incremental improvement over existing methods.
The authors tackled the problem of dialogue summarization not addressing user-specific interests by introducing instruction-tuning and a method to synthesize query-based summarization triples, resulting in a model that outperforms state-of-the-art models and shows higher generalizability and faithfulness.
Conventional dialogue summarization methods directly generate summaries and do not consider user's specific interests. This poses challenges in cases where the users are more focused on particular topics or aspects. With the advancement of instruction-finetuned language models, we introduce instruction-tuning to dialogues to expand the capability set of dialogue summarization models. To overcome the scarcity of instructive dialogue summarization data, we propose a three-step approach to synthesize high-quality query-based summarization triples. This process involves summary-anchored query generation, query filtering, and query-based summary generation. By training a unified model called InstructDS (Instructive Dialogue Summarization) on three summarization datasets with multi-purpose instructive triples, we expand the capability of dialogue summarization models. We evaluate our method on four datasets, including dialogue summarization and dialogue reading comprehension. Experimental results show that our approach outperforms the state-of-the-art models and even models with larger sizes. Additionally, our model exhibits higher generalizability and faithfulness, as confirmed by human subjective evaluations.