Explore, Propose, and Assemble: An Interpretable Model for Multi-Hop Reading Comprehension
This addresses the problem of interpretable reasoning in multi-hop reading comprehension for AI systems, though it is incremental as it builds on existing methods with a novel modular approach.
The paper tackles multi-hop reading comprehension by proposing an interpretable three-module system (EPAr) that explores, proposes, and assembles answers from multiple documents, achieving significant improvements over baselines and competitive results on WikiHop and MedHop datasets.
Multi-hop reading comprehension requires the model to explore and connect relevant information from multiple sentences/documents in order to answer the question about the context. To achieve this, we propose an interpretable 3-module system called Explore-Propose-Assemble reader (EPAr). First, the Document Explorer iteratively selects relevant documents and represents divergent reasoning chains in a tree structure so as to allow assimilating information from all chains. The Answer Proposer then proposes an answer from every root-to-leaf path in the reasoning tree. Finally, the Evidence Assembler extracts a key sentence containing the proposed answer from every path and combines them to predict the final answer. Intuitively, EPAr approximates the coarse-to-fine-grained comprehension behavior of human readers when facing multiple long documents. We jointly optimize our 3 modules by minimizing the sum of losses from each stage conditioned on the previous stage's output. On two multi-hop reading comprehension datasets WikiHop and MedHop, our EPAr model achieves significant improvements over the baseline and competitive results compared to the state-of-the-art model. We also present multiple reasoning-chain-recovery tests and ablation studies to demonstrate our system's ability to perform interpretable and accurate reasoning.