Enhancing Large Language Model with Decomposed Reasoning for Emotion Cause Pair Extraction

TL;DR

This paper introduces the DECC framework that leverages chain-of-thought prompting, inference, and pruning to improve large language model performance on emotion-cause pair extraction, surpassing supervised methods.

Contribution

It proposes a novel decomposed reasoning framework with logical pruning and in-context learning to enhance LLMs for ECPE without additional training.

Findings

DECC outperforms state-of-the-art supervised fine-tuning methods.

The framework is effective across different LLMs and datasets.

Component analysis shows the robustness of the approach.

Abstract

Emotion-Cause Pair Extraction (ECPE) involves extracting clause pairs representing emotions and their causes in a document. Existing methods tend to overfit spurious correlations, such as positional bias in existing benchmark datasets, rather than capturing semantic features. Inspired by recent work, we explore leveraging large language model (LLM) to address ECPE task without additional training. Despite strong capabilities, LLMs suffer from uncontrollable outputs, resulting in mediocre performance. To address this, we introduce chain-of-thought to mimic human cognitive process and propose the Decomposed Emotion-Cause Chain (DECC) framework. Combining inducing inference and logical pruning, DECC guides LLMs to tackle ECPE task. We further enhance the framework by incorporating in-context learning. Experiment results demonstrate the strength of DECC compared to state-of-the-art supervised fine-tuning methods. Finally, we analyze the effectiveness of each component and the robustness of the method in various scenarios, including different LLM bases, rebalanced datasets, and multi-pair extraction.