Controllable Abstractive Dialogue Summarization with Sketch Supervision

Abstract

In this paper, we aim to improve abstractive dialogue summarization quality and, at the same time, enable granularity control. Our model has two primary components and stages: 1) a two-stage generation strategy that generates a preliminary summary sketch serving as the basis for the final summary. This summary sketch provides a weakly supervised signal in the form of pseudo-labeled interrogative pronoun categories and key phrases extracted using a constituency parser. 2) A simple strategy to control the granularity of the final summary, in that our model can automatically determine or control the number of generated summary sentences for a given dialogue by predicting and highlighting different text spans from the source text. Our model achieves state-of-the-art performance on the largest dialogue summarization corpus SAMSum, with as high as 50.79 in ROUGE-L score. In addition, we conduct a case study and show competitive human evaluation results and controllability to human-annotated summaries.

arXiv commentACL-Findings 2021. Code is released at https://github.com/salesforce/ConvSumm

References in corpus (20)

• Adam: A Method for Stochastic Optimization
• Teaching Machines to Read and Comprehend
• BERTScore: Evaluating Text Generation with BERT
• A Deep Reinforced Model for Abstractive Summarization
• PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization
• CTRL: A Conditional Transformer Language Model for Controllable Generation
• Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond
• SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarization
• TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agents
• Get To The Point: Summarization with Pointer-Generator Networks
• DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation
• Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting
• Constituency Parsing with a Self-Attentive Encoder
• Scientific Paper Summarization Using Citation Summary Networks
• Data-driven Summarization of Scientific Articles
• Length-controllable Abstractive Summarization by Guiding with Summary Prototype
• Incorporating Commonsense Knowledge into Abstractive Dialogue Summarization via Heterogeneous Graph Networks
• MoverScore: Text Generation Evaluating with Contextualized Embeddings and Earth Mover Distance
• Dial2Desc: End-to-end Dialogue Description Generation
• Neural Latent Extractive Document Summarization