Multi-hop Reading Comprehension through Question Decomposition and Rescoring

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 6097–6109 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 6097 Multi-hop Reading Comprehension through Question Decomposition and Rescoring Sewon Min1, Victor Zhong1, Luke Zettlemoyer1, Hannaneh Hajishirzi1,2 1University of Washington 2Allen Institute for Artificial Intelligence {sewon,vzhong,lsz,hannaneh}@cs.washington.edu Abstract Multi-hop Reading Comprehension (RC) re- quires reasoning and aggregation across sev- eral paragraphs. We propose a system for multi-hop RC that decomposes a composi- tional question into simpler sub-questions that can be answered by off-the-shelf single-hop RC models. Since annotations for such de- composition are expensive, we recast sub- question generation as a span prediction prob- lem and show that our method, trained us- ing only 400 labeled examples, generates sub-questions that are as effective as human- authored sub-questions. We also introduce a new global rescoring approach that considers each decomposition (i.e. the sub-questions and their answers) to select the best final answer, greatly improving overall performance. Our experiments on HOTPOTQA show that this approach achieves the state-of-the-art results, while providing explainable evidence for its decision making in the form of sub-questions. 1 Introduction Multi-hop reading comprehension (RC) is chal- lenging because it requires the aggregation of evi- dence across several paragraphs to answer a ques- tion. Table 1 shows an example of multi-hop RC, where the question “Which team does the player named 2015 Diamond Head Classics MVP play for?” requires first finding the player who won MVP from one paragraph, and then finding the team that player plays for from another paragraph. In this paper, we propose DECOMPRC, a sys- tem for multi-hop RC, that learns to break compo- sitional multi-hop questions into simpler, single- hop sub-questions using spans from the original question. For example, for the question in Ta- ble 1, we can create the sub-questions “Which player named 2015 Diamond Head Classics MVP?” and “Which team does ANS play for?”, Q Which team does the player named 2015 Diamond Head Classics MVP play for? P1 The 2015 Diamond Head Classic was ... Buddy Hield was named the tournament’s MVP. P2 Chavano Rainier Buddy Hield is a Bahamian professional basketball player for the Sacramento Kings ... Q1 Which player named 2015 Diamond Head Classics MVP? Q2 Which team does ANS play for? Table 1: An example of multi-hop question from HOT- POTQA. The first cell shows given question and two of given paragraphs (other eight paragraphs are not shown), where the red text is the groundtruth answer. Our system selects a span over the question and writes two sub-questions shown in the second cell. where the token ANS is replaced by the answer to the first sub-question. The final answer is then the answer to the second sub-question. Recent work on question decomposition relies on distant supervision data created on top of un- derlying relational logical forms (Talmor and Be- rant, 2018), making it difficult to generalize to diverse natural language questions such as those on HOTPOTQA (Yang et al., 2018). In contrast, …

Content truncated to 500 words