Process Rewards with Learned Reliability

TL;DR

BetaPRM introduces a distributional reward model that predicts both success probabilities and reliability, enabling more trustworthy step-level feedback and improving reasoning efficiency.

Contribution

The paper proposes BetaPRM, a novel distributional PRM that learns a reliability signal for step rewards, enhancing decision trustworthiness and enabling adaptive computation strategies.

Findings

BetaPRM improves Best-of-N reasoning accuracy across benchmarks.

ACA reduces token usage by up to 33.57% while increasing accuracy.

BetaPRM maintains effective error detection with reliability signals.

Abstract

Process Reward Models (PRMs) provide step-level feedback for reasoning, but current PRMs usually output only a single reward score for each step. Downstream methods must therefore treat imperfect step-level reward predictions as reliable decision signals, with no indication of when these predictions should be trusted. We propose BetaPRM, a distributional PRM that predicts both a step-level success probability and the reliability of that prediction. Given step-success supervision from Monte Carlo continuations, BetaPRM learns a Beta belief that explains the observed number of successful continuations through a Beta-Binomial likelihood, rather than regressing to the finite-sample success ratio as a point target. This learned reliability signal indicates when a step reward should be trusted, enabling downstream applications to distinguish reliable rewards from uncertain ones. As one application, we introduce Adaptive Computation Allocation (ACA) for PRM-guided Best-of-N reasoning. ACA uses the learned reliability signal to stop when a high-reward solution is reliable and to spend additional computation on uncertain candidate prefixes. Experiments across four backbones and four reasoning benchmarks show that BetaPRM improves PRM-guided Best-of-N selection while preserving standard step-level error detection. Built on this signal, ACA improves the accuracy--token tradeoff over fixed-budget Best-of-16, reducing token usage by up to 33.57% while improving final-answer accuracy.