FC-KBQA: A Fine-to-Coarse Composition Framework for Knowledge Base Question Answering
It addresses generalization and executability problems in KBQA, offering a novel method for improving question answering over knowledge bases.
The paper tackles the generalization and executability issues in knowledge base question answering by proposing a Fine-to-Coarse Composition framework (FC-KBQA), which achieves state-of-the-art performance on GrailQA and WebQSP datasets and runs 4 times faster than the baseline.
The generalization problem on KBQA has drawn considerable attention. Existing research suffers from the generalization issue brought by the entanglement in the coarse-grained modeling of the logical expression, or inexecutability issues due to the fine-grained modeling of disconnected classes and relations in real KBs. We propose a Fine-to-Coarse Composition framework for KBQA (FC-KBQA) to both ensure the generalization ability and executability of the logical expression. The main idea of FC-KBQA is to extract relevant fine-grained knowledge components from KB and reformulate them into middle-grained knowledge pairs for generating the final logical expressions. FC-KBQA derives new state-of-the-art performance on GrailQA and WebQSP, and runs 4 times faster than the baseline.