Translucent Answer Predictions in Multi-Hop Reading Comprehension

Authors

  • G P Shrivatsa Bhargav IISc, Bangalore
  • Michael Glass IBM Research AI
  • Dinesh Garg IBM Research AI
  • Shirish Shevade IISc, Bangalore
  • Saswati Dana IBM Research AI
  • Dinesh Khandelwal IBM Research AI
  • L Venkata Subramaniam IBM Research AI
  • Alfio Gliozzo IBM Research AI

DOI:

https://doi.org/10.1609/aaai.v34i05.6272

Abstract

Research on the task of Reading Comprehension style Question Answering (RCQA) has gained momentum in recent years due to the emergence of human annotated datasets and associated leaderboards, for example CoQA, HotpotQA, SQuAD, TriviaQA, etc. While state-of-the-art has advanced considerably, there is still ample opportunity to advance it further on some important variants of the RCQA task. In this paper, we propose a novel deep neural architecture, called TAP (Translucent Answer Prediction), to identify answers and evidence (in the form of supporting facts) in an RCQA task requiring multi-hop reasoning. TAP comprises two loosely coupled networks – Local and Global Interaction eXtractor (LoGIX) and Answer Predictor (AP). LoGIX predicts supporting facts, whereas AP consumes these predicted supporting facts to predict the answer span. The novel design of LoGIX is inspired by two key design desiderata – local context and global interaction– that we identified by analyzing examples of multi-hop RCQA task. The loose coupling between LoGIX and the AP reveals the set of sentences used by the AP in predicting an answer. Therefore, answer predictions of TAP can be interpreted in a translucent manner. TAP offers state-of-the-art performance on the HotpotQA (Yang et al. 2018) dataset – an apt dataset for multi-hop RCQA task – as it occupies Rank-1 on its leaderboard (https://hotpotqa.github.io/) at the time of submission.

Downloads

Published

2020-04-03

How to Cite

Bhargav, G. P. S., Glass, M., Garg, D., Shevade, S., Dana, S., Khandelwal, D., Subramaniam, L. V., & Gliozzo, A. (2020). Translucent Answer Predictions in Multi-Hop Reading Comprehension. Proceedings of the AAAI Conference on Artificial Intelligence, 34(05), 7700-7707. https://doi.org/10.1609/aaai.v34i05.6272

Issue

Section

AAAI Technical Track: Natural Language Processing