Revealing and Mitigating the Challenge of Detecting Character Knowledge Errors in LLM Role-Playing
This addresses the challenge of ensuring authentic character knowledge for realistic LLM role-playing agents, which is incremental as it builds on existing work by focusing on a previously overlooked aspect.
The paper tackles the problem of detecting character knowledge errors in LLM role-playing, finding that even advanced LLMs struggle with this task, especially for familiar knowledge, and proposes a method that improves error detection but does not fully resolve the issue.
Large language model (LLM) role-playing has gained widespread attention. Authentic character knowledge is crucial for constructing realistic LLM role-playing agents. However, existing works usually overlook the exploration of LLMs' ability to detect characters' known knowledge errors (KKE) and unknown knowledge errors (UKE) while playing roles, which would lead to low-quality automatic construction of character trainable corpus. In this paper, we propose RoleKE-Bench to evaluate LLMs' ability to detect errors in KKE and UKE. The results indicate that even the latest LLMs struggle to detect these two types of errors effectively, especially when it comes to familiar knowledge. We experimented with various reasoning strategies and propose an agent-based reasoning method, Self-Recollection and Self-Doubt (S$^2$RD), to explore further the potential for improving error detection capabilities. Experiments show that our method effectively improves the LLMs' ability to detect error character knowledge, but it remains an issue that requires ongoing attention.