PreWoMe: Exploiting Presuppositions as Working Memory for Long Form Question Answering
This addresses the challenge of unpredictable, ambiguous questions in real-world QA settings, though it appears incremental as it builds on existing presupposition-based methods.
The paper tackles the problem of misleading questions in long-form question answering by proposing PreWoMe, which extracts presuppositions as working memory to generate feedback and action, showing effectiveness in handling both misleading and normal questions.
Information-seeking questions in long-form question answering (LFQA) often prove misleading due to ambiguity or false presupposition in the question. While many existing approaches handle misleading questions, they are tailored to limited questions, which are insufficient in a real-world setting with unpredictable input characteristics. In this work, we propose PreWoMe, a unified approach capable of handling any type of information-seeking question. The key idea of PreWoMe involves extracting presuppositions in the question and exploiting them as working memory to generate feedback and action about the question. Our experiment shows that PreWoMe is effective not only in tackling misleading questions but also in handling normal ones, thereby demonstrating the effectiveness of leveraging presuppositions, feedback, and action for real-world QA settings.