PreSTO Accelerates LLM Power Sampling via Subtree Prefetching
October 8, 2026
PreSTO accelerates the PowerMH sampler by prefetching future accept/reject decisions in the Metropolis-Hastings transition kernel. By batching proposals from a prefetched subtree, it improves test-time reasoning speed while maintaining the integrity of the original transition kernel.
HOW THIS AFFECTS YOU
●
builderYou can reduce the latency of power sampling for improved test-time LLM reasoning.
●
researcherYou can speed up Metropolis-Hastings updates without altering the underlying transition kernel.