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

Anonymous

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 \emph{Decomposed Emotion-Cause Chain (DECC)} framework. Combining inducing inference and logical pruning, DECC effectively guides LLMs to tackle ECPE task. We further enhance the framework by incorporating in-context learning. Experiment results demonstrate DECC's strength compared to state-of-the-art supervised fine-tuning methods. Finally we analyses each component's effectiveness and method’s robustness in various scenarios including different LLM base, rebalanced datasets, and multi-pair extraction.

Paper Type: long

Research Area: Sentiment Analysis, Stylistic Analysis, and Argument Mining

Contribution Types: NLP engineering experiment

Languages Studied: English, Chinese