MAQInstruct: Instruction-based Unified Event Relation Extraction

TL;DR

MAQInstruct introduces a novel instruction-based framework for event relation extraction that reduces sample complexity and improves performance across multiple large language models by transforming the task and incorporating bipartite matching loss.

Contribution

It presents a new approach transforming event relation extraction into event selection with relation instructions and uses bipartite matching loss to enhance instruction-based methods.

Findings

Significant performance improvements across multiple LLMs.

Reduced inference sample requirements.

Enhanced robustness of event relation extraction.

Abstract

Extracting event relations that deviate from known schemas has proven challenging for previous methods based on multi-class classification, MASK prediction, or prototype matching. Recent advancements in large language models have shown impressive performance through instruction tuning. Nevertheless, in the task of event relation extraction, instruction-based methods face several challenges: there are a vast number of inference samples, and the relations between events are non-sequential. To tackle these challenges, we present an improved instruction-based event relation extraction framework named MAQInstruct. Firstly, we transform the task from extracting event relations using given event-event instructions to selecting events using given event-relation instructions, which reduces the number of samples required for inference. Then, by incorporating a bipartite matching loss, we reduce the dependency of the instruction-based method on the generation sequence. Our experimental results demonstrate that MAQInstruct significantly improves the performance of event relation extraction across multiple LLMs.