Chinese Essay Rhetoric Recognition Using LoRA, In-context Learning and Model Ensemble

TL;DR

This paper employs LoRA fine-tuning, in-context learning, and model ensemble techniques with LLMs to improve Chinese essay rhetoric recognition, achieving top results in a competitive evaluation.

Contribution

It introduces a novel combination of LoRA, in-context learning, and ensemble methods for Chinese rhetoric recognition using LLMs, setting new state-of-the-art performance.

Findings

Achieved the best performance on all three tracks of CCL 2025 Chinese essay rhetoric recognition evaluation.

Utilized JSON-formatted outputs with Chinese key translation for structured results.

Won the first prize in the evaluation competition.

Abstract

Rhetoric recognition is a critical component in automated essay scoring. By identifying rhetorical elements in student writing, AI systems can better assess linguistic and higher-order thinking skills, making it an essential task in the area of AI for education. In this paper, we leverage Large Language Models (LLMs) for the Chinese rhetoric recognition task. Specifically, we explore Low-Rank Adaptation (LoRA) based fine-tuning and in-context learning to integrate rhetoric knowledge into LLMs. We formulate the outputs as JSON to obtain structural outputs and translate keys to Chinese. To further enhance the performance, we also investigate several model ensemble methods. Our method achieves the best performance on all three tracks of CCL 2025 Chinese essay rhetoric recognition evaluation task, winning the first prize.