主题论文总结2：会议/对话摘要任务（持续更新ing...）

诸神缄默不语-个人优快云博文目录

对话内容常常围绕主题展开。

1. 医疗领域对话摘要：
   1. Topic-aware Pointer-Generator Networks for Summarizing Spoken Conversations（第一个将pointer generator network¹用于对话摘要的论文）
   2. Dr.Summarize: Global Summarization of Medical Dialogue by Exploiting Local Structures
2. 法律领域对话摘要：
   1. Legal Summarization for Multi-role Debate Dialogue via Controversy Focus Mining and Multi-task Learning
3. How to Interact and Change? Abstractive Dialogue Summarization with Dialogue Act Weight and Topic Change Info
4. Improving Abstractive Dialogue Summarization with Hierarchical Pretraining and Topic Segment
5. Topic-Aware Contrastive Learning for Abstractive Dialogue Summarization
6. Unsupervised Summarization for Chat Logs with Topic-Oriented Ranking and Context-Aware Auto-Encoders
7. Multi-View Sequence-to-Sequence Models with Conversational Structure for Abstractive Dialogue Summarization
8. Topic-Oriented Spoken Dialogue Summarization for Customer Service with Saliency-Aware Topic Modeling
9. Extracting Decisions from Multi-Party Dialogue Using Directed Graphical Models and Semantic Similarity
10. Automatic analysis of multiparty meetings
11. Improving Abstractive Dialogue Summarization with Graph Structures and Topic Words
12. A Finer-grain Universal Dialogue Semantic Structures based Model For Abstractive Dialogue Summarization
13. Situation-Based Multiparticipant Chat Summarization: a Concept, an Exploration-Annotation Tool and an Example Collection
14. Other Roles Matter! Enhancing Role-Oriented Dialogue Summarization via Role Interactions
15. Summ^N: A Multi-Stage Summarization Framework for Long Input Dialogues and Documents
16. Dialogue Summaries as Dialogue States (DS2), Template-Guided Summarization for Few-shot Dialogue State Tracking
17. DialogLM: Pre-Trained Model for Long Dialogue Understanding and Summarization
18. A Survey on Dialogue Summarization: Recent Advances and New Frontiers：2022年综述
19. CONFIT: Toward Faithful Dialogue Summarization with Linguistically-Informed Contrastive Fine-tuning
20. Domain-Oriented Prefix-Tuning: Towards Efficient and Generalizable Fine-tuning for Zero-Shot Dialogue Summarization
21. From spoken dialogue to formal summary: An utterance rewriting for dialogue summarization：还没有发出来
22. TANet: Thread-Aware Pretraining for Abstractive Conversational Summarization
23. (2021 SIGIR 美团) Distant Supervision based Machine Reading Comprehension for Extractive Summarization in Customer Service
    美团官方讲解博文：对话摘要技术在美团的探索（SIGIR）
24. 会议摘要
    1. Automatic meeting summarization and topic detection system
    2. Keep Meeting Summaries on Topic: Abstractive Multi-Modal Meeting Summarization
    3. Incremental temporal summarization in multiparty meetings
    4. A template-based abstractive meeting summarization: Leveraging summary and source text relationships
25. 对话摘要指标评估：DialSummEval: Revisiting Summarization Evaluation for Dialogues

证明对话摘要任务适宜于生成式摘要而非抽取式摘要：Topic-aware Pointer-Generator Networks for Summarizing Spoken Conversations 一文的摘要。

重要公开数据集：

1. AMI Meeting Corpus（dialogue acts, topic descriptions, named entities, hand gestures, and gaze direction都是标注好的）
   官网：AMI Corpus - Annotation
   使用这一数据集的论文：
   Automatic analysis of multiparty meetings
2. SamSum
   使用这一数据集的论文：
   Multi-View Sequence-to-Sequence Models with Conversational Structure for Abstractive Dialogue Summarization
3. 其他整理数据集的项目：
   1. xcfcode/meeting_summarization_dataset: Codes for processing meeting summarization datasets AMI and ICSI.
   2. 对话摘要 | 抽取式与生成式 | 数据集与baseline_codebrid的博客-优快云博客_抽取式和生成式

重要术语：dialogue act https://en.wikipedia.org/wiki/Dialog_act
代表重要外部信息的interactive pattern，the effect of an utterance on the context，includes some conversational interactions and provides important information to understand dialogue
使用这一概念的论文：
How to Interact and Change? Abstractive Dialogue Summarization with Dialogue Act Weight and Topic Change Info
Extracting Decisions from Multi-Party Dialogue Using Directed Graphical Models and Semantic Similarity
Unsupervised Modeling of Twitter Conversations

其他整理会议/对话摘要论文的工作：

1. 对话文本摘要概述_海晨威的博客-优快云博客_文本摘要的应用场景

---

1. Re46：读论文PGN/Pointer-Generator Netwoks Get To The Point：Summarization with Pointer-Generator Networks ↩︎