# Talk Less, Call Right: Enhancing Role-Play LLM Agents with Automatic Prompt Optimization and Role Prompting

## TL;DR

This paper presents a rule-based role prompting method that significantly improves the performance of role-playing dialogue agents by enforcing function calls and designing effective prompts, outperforming other prompting strategies.

## Contribution

The study introduces a novel rule-based role prompting approach with character-card design and function call enforcement, enhancing dialogue agent reliability and effectiveness.

## Key findings

- RRP achieved a score of 0.571, surpassing the baseline of 0.519.
- RRP outperformed other prompting methods like APO.
- Open-sourced prompts and APO tool for future research.

## Abstract

This report investigates approaches for prompting a tool-augmented large language model (LLM) to act as a role-playing dialogue agent in the API track of the Commonsense Persona-grounded Dialogue Challenge (CPDC) 2025. In this setting, dialogue agents often produce overly long in-character responses (over-speaking) while failing to use tools effectively according to the persona (under-acting), such as generating function calls that do not exist or making unnecessary tool calls before answering. We explore four prompting approaches to address these issues: 1) basic role prompting, 2) improved role prompting, 3) automatic prompt optimization (APO), and 4) rule-based role prompting. The rule-based role prompting (RRP) approach achieved the best performance through two novel techniques-character-card/scene-contract design and strict enforcement of function calling-which led to an overall score of 0.571, improving on the zero-shot baseline score of 0.519. These findings demonstrate that RRP design can substantially improve the effectiveness and reliability of role-playing dialogue agents compared with more elaborate methods such as APO. To support future efforts in developing persona prompts, we are open-sourcing all of our best-performing prompts and the APO tool Source code is available at https://github.com/scb-10x/apo