Event-Centric Question Answering via Contrastive Learning and Invertible Event Transformation

TL;DR

This paper introduces TranCLR, a novel event-centric question answering model that uses contrastive learning and invertible event transformation to improve understanding of event relations, significantly enhancing answer accuracy.

Contribution

The paper proposes a new QA model with an invertible transformation and contrastive learning to inject event semantic knowledge into QA systems.

Findings

Over 8.4% gain in token-level F1 score

3.0% improvement in Exact Match score

Effective injection of event knowledge into QA models

Abstract

Human reading comprehension often requires reasoning of event semantic relations in narratives, represented by Event-centric Question-Answering (QA). To address event-centric QA, we propose a novel QA model with contrastive learning and invertible event transformation, call TranCLR. Our proposed model utilizes an invertible transformation matrix to project semantic vectors of events into a common event embedding space, trained with contrastive learning, and thus naturally inject event semantic knowledge into mainstream QA pipelines. The transformation matrix is fine-tuned with the annotated event relation types between events that occurred in questions and those in answers, using event-aware question vectors. Experimental results on the Event Semantic Relation Reasoning (ESTER) dataset show significant improvements in both generative and extractive settings compared to the existing strong baselines, achieving over 8.4% gain in the token-level F1 score and 3.0% gain in Exact Match (EM) score under the multi-answer setting. Qualitative analysis reveals the high quality of the generated answers by TranCLR, demonstrating the feasibility of injecting event knowledge into QA model learning. Our code and models can be found at https://github.com/LuJunru/TranCLR.