基于主题感知话语表示的无监督对话主题分割
Unsupervised Dialogue Topic Segmentation with Topic-aware Utterance Representation
AI总结：
针对现有无监督对话主题分割方法主题识别不充分、未标注数据利用不足的问题，提出通过邻接话语匹配与伪分割学习主题感知话语表示的框架，在两个基准上性能优于强基线。
摘要：
对话主题分割（Dialogue Topic Segmentation，DTS）在各类对话建模任务中发挥着核心作用。现有无监督DTS方法要么依托语义相似度，要么依托对话连贯性来评估主题相似度，以此实现分割。但仅通过语义相似度或对话连贯性无法完全识别主题相似性，且蕴含话语关系有用线索的未标注对话数据也未得到充分利用。本文提出一种新型无监督DTS框架，通过邻接话语匹配与伪分割从未标注对话数据中学习主题感知话语表示。在DialSeg711和Doc2Dial两个基准数据集上开展的大量实验表明，该方法的性能显著优于强基线方法。为保证可复现性，我们的代码与数据发布于：https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/dial-start。
英文摘要：
Dialogue Topic Segmentation (DTS) plays an essential role in a variety of dialogue modeling tasks. Previous DTS methods either focus on semantic similarity or dialogue coherence to assess topic similarity for unsupervised dialogue segmentation. However, the topic similarity cannot be fully identified via semantic similarity or dialogue coherence. In addition, the unlabeled dialogue data, which contains useful clues of utterance relationships, remains underexploited. In this paper, we propose a novel unsupervised DTS framework, which learns topic-aware utterance representations from unlabeled dialogue data through neighboring utterance matching and pseudo-segmentation. Extensive experiments on two benchmark datasets (i.e., DialSeg711 and Doc2Dial) demonstrate that our method significantly outperforms the strong baseline methods. For reproducibility, we provide our code and data at:https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/dial-start.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
