Process Rewards with Learned Reliability
Quick Answer
BetaPRM introduces a distributional approach to Process Reward Models, predicting both success probability and reliability, enhancing decision-making in AI systems.
Quick Take
It enables Adaptive Computation Allocation, improving accuracy-token tradeoff by up to 33.57% while maintaining error detection across four benchmarks.
Key Points
- BetaPRM predicts step-level success probability and reliability for better decision signals.
- Adaptive Computation Allocation (ACA) optimizes computation based on reward reliability.
- Experiments show up to 33.57% reduction in token usage while improving accuracy.
- BetaPRM maintains standard error detection across four reasoning benchmarks.
- Model improves PRM-guided Best-of-N selection significantly.
Paper Resources
📖 Reader Mode
~2 min readAbstract: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.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.15529 [cs.CL] |
| (or arXiv:2605.15529v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.15529 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jinyuan Li [view email]
[v1]
Fri, 15 May 2026 01:57:11 UTC (4,966 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.