Flipping Against All Odds: Reducing LLM Coin Flip Bias via Verbalized Rejection Sampling

ArXi:2506.09998v2 Announce Type: replace Large language models (LLMs) can often accurately describe probability distributions using natural language, yet they still struggle to generate faithful samples from them. This mismatch limits their use in tasks requiring reliable stochasticity, such as Monte Carlo methods, agent-based simulations, and randomized decision-making. We investigate this gap between knowledge and sampling in the context of Bernoulli distributions. We