POKERBENCH: Training Large Language Models to become Professional Poker Players

In this Opik Virtual Learning Series session, a live paper discussion of “POKERBENCH: Training Large Language Models to become Professional Poker Players” Claire Longo and Abby Morgan are joined by the paper authors Richard Zhuang, Akshat Gupta, and Brian Li to break down complex concepts and highlight key takeaways from their research. Learn why poker is a perfect test case for LLM reasoning capabilities, and how the authors approached fine-tuning and testing models on a specialized domain.