Answering Unanswered Questions through Semantic Reformulations in Spoken QA
This addresses user experience issues in voice assistants by reducing unanswered questions, though it is incremental as it builds on existing QA systems.
The paper tackled the problem of unanswered questions in spoken QA due to disfluencies and informal phrasing, proposing a semantic reformulation model that improved answer rates by up to 24% on previously unanswered questions with 75% relevance.
Spoken Question Answering (QA) is a key feature of voice assistants, usually backed by multiple QA systems. Users ask questions via spontaneous speech which can contain disfluencies, errors, and informal syntax or phrasing. This is a major challenge in QA, causing unanswered questions or irrelevant answers, and leading to bad user experiences. We analyze failed QA requests to identify core challenges: lexical gaps, proposition types, complex syntactic structure, and high specificity. We propose a Semantic Question Reformulation (SURF) model offering three linguistically-grounded operations (repair, syntactic reshaping, generalization) to rewrite questions to facilitate answering. Offline evaluation on 1M unanswered questions from a leading voice assistant shows that SURF significantly improves answer rates: up to 24% of previously unanswered questions obtain relevant answers (75%). Live deployment shows positive impact for millions of customers with unanswered questions; explicit relevance feedback shows high user satisfaction.