Improving Machine Reading Comprehension with Contextualized Commonsense Knowledge

To perform well on a machine reading comprehension (MRC) task, machine readers usually require commonsense knowledge that is not explicitly mentioned in the given documents. This paper aims to extract a new kind of structured knowledge from scripts and use it to improve MRC. We focus on scripts as they contain rich verbal and nonverbal messages, and two relevant messages originally conveyed by different modalities during a short time period may serve as arguments of a piece of commonsense knowledge as they function together in daily communications. To save human efforts to name relations, we propose to represent relations implicitly by situating such an argument pair in a context and call it contextualized knowledge.

To use the extracted knowledge to improve MRC, we compare several fine-tuning strategies to use the weakly-labeled MRC data constructed based on contextualized knowledge and further design a teacher-student paradigm with multiple teachers to facilitate the transfer of knowledge in weakly-labeled MRC data. Experimental results show that our paradigm outperforms other methods that use weaklylabeled data and improves a state-of-the-art baseline by 4.3% in accuracy on a Chinese multiple-choice MRC dataset C 3 , wherein most of the questions require unstated prior knowledge. We also seek to transfer the knowledge to other tasks by simply adapting the resulting student reader, yielding a 2.9% improvement in F1 on a relation extraction dataset DialogRE, demonstrating the potential usefulness of the knowledge for non-MRC tasks that require document comprehension. Interior. Runaway office. Day. Andy: I tried to ask her, but... Emily: You never ask Miranda. Anything. (sighs) All right, I'll take care of the other stuff. You go to Calvin Klein. Andy: Me? Emily: I'm sorry. Do you have a prior commitment? Is there some hideous pants convention? Andy: So I just, what, go down to the Calvin Klein store and ask them... 3 Emily rolls her eyes so hard they almost eject from her head. Emily: You're not going to the store. Andy: Of course not. I'm going...(thinking)...to his house. Emily (oh god): You are catching on quickly. We always send assistants to a designer's home on their very first day. You're going to his showroom. I'll give you the address. Andy: Sorry. Got it. What's the nearest subway stop? Emily: Good God. You do not. Under any circumstances. Take public transportation. Andy: I don't?