-
Notifications
You must be signed in to change notification settings - Fork 27
Description
Hello,
First of all, thank you so much for creating and maintaining this repository! 🙏
It has been extremely helpful for my research, and I wouldn’t have been able to publish my paper as smoothly without it.
I’m writing to kindly request consideration for adding our recent work to the list:
MIMIC: Integrating Diverse Personality Traits for Better Game Testing Using Large Language Models
Accepted at the 40th IEEE/ACM International Conference on Automated Software Engineering (ASE 2025)
• Paper: https://arxiv.org/abs/2510.01635
• Homepage: https://mimic-persona.github.io/MIMIC-Home-Page/
• Code: https://github.com/Mimic-Persona/MIMIC
Game Types
Our agent can run across three different games:
- Minecraft – a game you already have a category for
- Shattered Pixel Dungeon – a widely played open-source, roguelike, turn-based RPG dungeon crawler
- Dungeon Adventures – an open-source, turn-based RPG dungeon crawler
Summary of Our Work
We introduce MIMIC, a novel framework that integrates diverse personality traits into game-testing agents.
Why does this matter?
- Traditional agents often optimize for the “best” solution, leading to repetitive strategies, BUT that’s NOT how humans play. Players bring different playstyles shaped by their personalities.
- MIMIC innovatively integrates personality into game testing with the help of LLMs, enabling richer in-game interactions, higher coverage, and more diverse solutions.
Key results:
In games like Minecraft, MIMIC outperforms state-of-the-art agents in task completion, coverage, and solution diversity. These results highlight the potential of personality-driven agents for software testing.
Thank you again for your consideration, and for maintaining this valuable resource!