A Discrete Hard EM Approach for Weakly Supervised Question Answering
EMNLP/IJCNLP (1), pp. 2851-2864, 2019.
Many question answering (QA) tasks only provide weak supervision for how the answer should be computed. For example, TriviaQA answers are entities that can be mentioned multiple times in supporting documents, while DROP answers can be computed by deriving many different equations from numbers in the reference text. In this paper, we sho...More
PPT (Upload PPT)