QAConv: Question Answering on Informative Conversations
This provides a new training and evaluation testbed for QA on conversations, addressing a domain-specific gap in handling informative dialogues like business emails and panel discussions.
The authors tackled the problem of question answering on long, complex informative conversations by introducing QAConv, a dataset of 34,608 QA pairs from 10,259 conversations, and found that state-of-the-art pretrained QA systems show limited zero-shot performance, often predicting questions as unanswerable.
This paper introduces QAConv, a new question answering (QA) dataset that uses conversations as a knowledge source. We focus on informative conversations, including business emails, panel discussions, and work channels. Unlike open-domain and task-oriented dialogues, these conversations are usually long, complex, asynchronous, and involve strong domain knowledge. In total, we collect 34,608 QA pairs from 10,259 selected conversations with both human-written and machine-generated questions. We use a question generator and a dialogue summarizer as auxiliary tools to collect and recommend questions. The dataset has two testing scenarios: chunk mode and full mode, depending on whether the grounded partial conversation is provided or retrieved. Experimental results show that state-of-the-art pretrained QA systems have limited zero-shot performance and tend to predict our questions as unanswerable. Our dataset provides a new training and evaluation testbed to facilitate QA on conversations research.