Evaluating Character Understanding of Large Language Models via Character Profiling from Fictional Works

TL;DR

This paper introduces a novel evaluation method for large language models' understanding of fictional characters through character profiling, demonstrating their capability to generate accurate profiles that are useful for role-playing agents.

Contribution

It proposes the character profiling task and the CroSS dataset to better assess LLMs' nuanced understanding of fictional characters, filling a gap in existing evaluation methods.

Findings

LLMs can generate accurate character profiles from fictional works.

Character profiling correlates well with downstream task performance.

The proposed method validates LLMs' nuanced character understanding.

Abstract

Large language models (LLMs) have demonstrated impressive performance and spurred numerous AI applications, in which role-playing agents (RPAs) are particularly popular, especially for fictional characters. The prerequisite for these RPAs lies in the capability of LLMs to understand characters from fictional works. Previous efforts have evaluated this capability via basic classification tasks or characteristic imitation, failing to capture the nuanced character understanding with LLMs. In this paper, we propose evaluating LLMs' character understanding capability via the character profiling task, i.e., summarizing character profiles from corresponding materials, a widely adopted yet understudied practice for RPA development. Specifically, we construct the CroSS dataset from literature experts and assess the generated profiles by comparing them with ground truth references and evaluating their applicability in downstream tasks. Our experiments, which cover various summarization methods and LLMs, have yielded promising results. These results strongly validate the character understanding capability of LLMs. Resources are available at https://github.com/Joanna0123/character_profiling.