2021/12/14 by Sara Brin Rosenthal, Sara Rosenthal, Mihaela Bornea +8 · 2 citations
Computer Science · Mathematics · #Boolean model #Computation and Language (cs.CL) #Computer science #Discrete mathematics #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Information retrieval #Linguistics #Mathematics #Philosophy #Programming language #Relevance (law) #Set (abstract data type) #Software Engineering Research #Topic Modeling #Word (group theory) #cs.CL
paper · pdf · doi:10.48550/arxiv.2112.07772
published in arXiv (Cornell University) (Cornell University) · 9 pages
arxiv created 2021/12/14 · openalex publication_date 2021/12/14 · arxiv updated 2021/12/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Existing datasets that contain boolean questions, such as BoolQ and TYDI QA , provide the user with a YES/NO response to the question. However, a one word response is not sufficient for an explainable system. We promote explainability by releasing a new set of annotations marking the evidence in existing TyDi QA and BoolQ datasets. We show that our annotations can be used to train a model that extracts improved evidence spans compared to models that rely on existing resources. We confirm our findings with a user study which shows that our extracted evidence spans enhance the user experience. We also provide further insight into the challenges of answering boolean questions, such as passages containing conflicting YES and NO answers, and varying degrees of relevance of the predicted evidence.