nicolay-r at SemEval-2024 Task 3: Using Flan-T5 for Reasoning Emotion Cause in Conversations with Chain-of-Thought on Emotion States.

Abstract

Emotion expression is one of the essential traits of conversations. It may be self-related or caused by another speaker. The variety of reasons may serve as a source of the further emotion causes: conversation history, speaker's emotional state, etc. Inspired by the most recent advances in Chain-of-Thought, in this work, we exploit the existing three-hop reasoning approach (THOR) to perform large language model instruction-tuning for answering: emotion states (THOR<inf>STATE</inf>), and emotion caused by one speaker to the other (THOR<inf>CAUSE</inf>). We equip THOR<inf>CAUSE</inf> with the reasoning revision (RR) for devising a reasoning path in fine-tuning. In particular, we rely on the annotated speaker emotion states to revise reasoning path. Our final submission, based on Flan-T5<inf>base</inf> (250M) and the rule-based span correction technique, preliminary tuned with THOR<inf>STATE</inf> and fine-tuned with THOR<inf>CAUSE-RR</inf> on competition training data, results in 3^rd and 4^th places (F1<inf>proportional</inf>) and 5^th place (F1<inf>strict</inf>) among 15 participating teams. Our THOR implementation fork is publicly available: https://github.com/nicolay-r/THOR-ECAC.