bayesian-persuasion

This paper introduces SDR-Bench, a benchmark for evaluating the personalization capabilities of large language models in a two-party Bayesian Persuasion framework, finding a consistent plateau across frontier LLMs and validating the framework with a field deployment.

This paper proposes a structured reinforcement learning framework for Bayesian persuasion in interactive driving, where a lead vehicle selectively reveals traffic information to guide connected vehicles. The method introduces MAPL and SQP algorithms, achieving 30% cost efficiency over existing methods.