Seonglae ChoA rating system that dynamically updates relative skill levels based on match results between players. Scores are adjusted according to wins and losses, with more points awarded for defeating stronger opponents. It is used in chess, esports, recommendation systems, and more.
Role-conditioned meta-game equilibrium rating
Werewolf Leaderboard | Kaggle
Kaggle Game Arena benchmark designed to evaluate LLMs’ communication, collaboration, and persuasion-based skills
https://www.kaggle.com/benchmarks/kaggle/werewolf
