Do I have the Knowledge to Answer? Investigating Answerability of Knowledge Base Questions
This addresses the issue of KB incompleteness for KBQA systems, which is an incremental step as it adapts an existing dataset to study unanswerability.
The authors tackled the problem of unanswerable questions in knowledge base question answering (KBQA) by creating GrailQAbility, a new benchmark dataset with unanswerability, and found that three state-of-the-art KBQA models experienced performance drops and struggled with specific forms of unanswerability.
When answering natural language questions over knowledge bases, missing facts, incomplete schema and limited scope naturally lead to many questions being unanswerable. While answerability has been explored in other QA settings, it has not been studied for QA over knowledge bases (KBQA). We create GrailQAbility, a new benchmark KBQA dataset with unanswerability, by first identifying various forms of KB incompleteness that make questions unanswerable, and then systematically adapting GrailQA (a popular KBQA dataset with only answerable questions). Experimenting with three state-of-the-art KBQA models, we find that all three models suffer a drop in performance even after suitable adaptation for unanswerable questions. In addition, these often detect unanswerability for wrong reasons and find specific forms of unanswerability particularly difficult to handle. This underscores the need for further research in making KBQA systems robust to unanswerability