RoleRMBench

RoleRMBench is a systematic benchmark for evaluating reward modeling in role-playing dialogue, encompassing seven fine-grained capabilities such as narrative management and role consistency.

Papers using RoleRMBench (3)

Persona-Pruner: Sculpting Lightweight Models for Role-Playing2026RoleRMBench & RoleRM: Towards Reward Modeling for Profile-Based Role Play in Dialogue Systems2025Beware of Your Po! Measuring and Mitigating AI Safety Risks in Role-Play Fine-Tuning of LLMs2025