Locate Then Ask: Interpretable Stepwise Reasoning for Multi-hop Question Answering
This addresses the need for more accurate and interpretable reasoning in multi-hop QA systems, though it is incremental as it builds on existing decomposition methods.
The paper tackles the problem of inaccurate decompositions in multi-hop question answering by proposing an interpretable stepwise reasoning framework that incorporates supporting sentence identification and question generation at each step, resulting in performance boosts on HotpotQA and 2WikiMultiHopQA datasets without decomposition supervision.
Multi-hop reasoning requires aggregating multiple documents to answer a complex question. Existing methods usually decompose the multi-hop question into simpler single-hop questions to solve the problem for illustrating the explainable reasoning process. However, they ignore grounding on the supporting facts of each reasoning step, which tends to generate inaccurate decompositions. In this paper, we propose an interpretable stepwise reasoning framework to incorporate both single-hop supporting sentence identification and single-hop question generation at each intermediate step, and utilize the inference of the current hop for the next until reasoning out the final result. We employ a unified reader model for both intermediate hop reasoning and final hop inference and adopt joint optimization for more accurate and robust multi-hop reasoning. We conduct experiments on two benchmark datasets HotpotQA and 2WikiMultiHopQA. The results show that our method can effectively boost performance and also yields a better interpretable reasoning process without decomposition supervision.