Process Supervision for Chain-of-Thought Reasoning via Monte Carlo Net Information Gain

• URL: http://arxiv.org/abs/2603.17815v1
• Date: Wed, 18 Mar 2026 15:08:01 GMT
• Title: Process Supervision for Chain-of-Thought Reasoning via Monte Carlo Net Information Gain
• Abstract summary: Multi-step reasoning improves the capabilities of large language models (LLMs) but increases the risk of errors propagating through intermediate steps.<n>Process reward models (PRMs) mitigate this by scoring each step individually, enabling fine-grained supervision and improved reliability.<n>We propose a novel approach to automatically generate step-level labels using Information Theory.
• Score: 23.290482237410398
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Multi-step reasoning improves the capabilities of large language models (LLMs) but increases the risk of errors propagating through intermediate steps. Process reward models (PRMs) mitigate this by scoring each step individually, enabling fine-grained supervision and improved reliability. Existing methods for training PRMs rely on costly human annotations or computationally intensive automatic labeling. We propose a novel approach to automatically generate step-level labels using Information Theory. Our method estimates how each reasoning step affects the likelihood of the correct answer, providing a signal of step quality. Importantly, it reduces computational complexity to $\mathcal{O}(N)$, improving over the previous $\mathcal{O}(N \log N)$ methods. We demonstrate that these labels enable effective chain-of-thought selection in best-of-$K$ evaluation settings across diverse reasoning benchmarks, including mathematics, Python programming, SQL, and scientific question answering. This work enables scalable and efficient supervision of LLM reasoning, particularly for tasks where error propagation is critical.

Related papers

• Unleashing Implicit Rewards: Prefix-Value Learning for Distribution-Level Optimization [74.91418266859297]
  Implicit process reward models (PRMs) provide fine-grained reward signals along the reasoning process.<n>PRMs suffer from a train-inference mismatch: training only constrains a sequence-level aggregate, whereas inference requires token-level scores to reflect local step quality.<n>We propose a novel Implicit Prefix-Value Reward Model (IPVRM), which directly learns a prefix-conditioned value function estimating the probability of eventual correctness.<n>We also propose Distribution-Level RL (DistRL), which computes TD advantages for both sampled tokens and high-probability candidate tokens.
  arXiv  Detail & Related papers  (2026-04-14T18:19:54Z)
• Efficient Process Reward Modeling via Contrastive Mutual Information [8.346479730119436]
  We propose contrastive pointwise mutual information (CPMI) as a novel automatic reward labeling method.<n>CPMI quantifies how much a reasoning step increases the mutual information between the step and the correct target answer.<n>We show that CPMI-based labeling reduces dataset construction time by 84% and token generation by 98% compared to MC estimation.
  arXiv  Detail & Related papers  (2026-04-12T14:28:35Z)
• Adversarial Training for Process Reward Models [47.92183495904245]
  We introduce Adversarially Trained PRMs (textttAPRM), where a Generator ($G$) learns to produce reasoning errors to deceive a PRM ($R$)<n>This interaction yields progressively harder negatives for $R$, improving its robustness and generalization to novel errors without requiring manual step-level labels.
  arXiv  Detail & Related papers  (2025-11-28T05:32:01Z)
• SPARE: Single-Pass Annotation with Reference-Guided Evaluation for Automatic Process Supervision and Reward Modelling [58.05959902776133]
  We introduce Single-Pass.<n>with Reference-Guided Evaluation (SPARE), a novel structured framework that enables efficient per-step annotation.<n>We demonstrate SPARE's effectiveness across four diverse datasets spanning mathematical reasoning (GSM8K, MATH), multi-hop question answering (MuSiQue-Ans), and spatial reasoning (SpaRP)<n>On ProcessBench, SPARE demonstrates data-efficient out-of-distribution generalization, using only $sim$16% of training samples compared to human-labeled and other synthetically trained baselines.
  arXiv  Detail & Related papers  (2025-06-18T14:37:59Z)
• Right Is Not Enough: The Pitfalls of Outcome Supervision in Training LLMs for Math Reasoning [35.142294794883455]
  We introduce MathOlympiadEval, a new dataset with fine-grained annotations, which reveals a significant gap between LLMs' answer correctness and their low process correctness.<n>Existing automated methods like LLM-as-a-judge struggle to reliably detect these reasoning flaws.<n>We propose ParaStepVerifier, a novel methodology for meticulous, step-by-step verification of mathematical solutions.
  arXiv  Detail & Related papers  (2025-06-07T17:54:56Z)
• Process Reward Models That Think [85.06022494911811]
  Step-by-step verifiers -- also known as process reward models (PRMs) -- are a key ingredient for test-time scaling.<n>This work aims to build data-efficient PRMs as verbalized step-wise reward models that verify every step in the solution by generating a verification chain-of-thought (CoT)<n>We propose ThinkPRM, a long CoT verifier fine-tuned on orders of magnitude fewer process labels than those required by discriminative PRMs.
  arXiv  Detail & Related papers  (2025-04-23T15:44:54Z)
• The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models [69.798277882245]
  We introduce Unsupervised Prefix Fine-Tuning (UPFT) to enhance large language models' reasoning efficiency.<n>UPFT removes the need for labeled data or exhaustive sampling.<n> Experiments show that UPFT matches the performance of supervised methods.
  arXiv  Detail & Related papers  (2025-03-04T18:56:03Z)
• Rational Metareasoning for Large Language Models [17.479428400594028]
  Being prompted to engage in reasoning has emerged as a core technique for using large language models (LLMs)<n>This work introduces a novel approach based on computational models of metareasoning used in cognitive science.<n>We develop a reward function that incorporates the Value of Computation by penalizing unnecessary reasoning.
  arXiv  Detail & Related papers  (2024-10-07T23:48:52Z)
• Improve Mathematical Reasoning in Language Models by Automated Process Supervision [23.807288360423193]
  We propose a novel divide-and-conquer style Monte Carlo Tree Search (MCTS) algorithm named textitOmegaPRM for the efficient collection of high-quality process supervision data.<n>We are able to collect over 1.5 million process supervision annotations to train Process Reward Models (PRMs)<n>This fully automated process supervision alongside the weighted self-consistency algorithm is able to enhance LLMs' math reasoning performances.
  arXiv  Detail & Related papers  (2024-06-05T19:25:40Z)
• Training Chain-of-Thought via Latent-Variable Inference [30.21067593018967]
  Large language models (LLMs) solve problems more accurately and interpretably when instructed to work out the answer step by step using a chain-of-thought'' prompt. Naively combining CoT with supervised tuning requires supervision not just of the correct answers, but also of detailed rationales that lead to those answers. We propose a fine-tuning strategy that tries to maximize the emphmarginal log-likelihood of generating a correct answer using CoT prompting.
  arXiv  Detail & Related papers  (2023-11-28T17:47:32Z)
• Guiding Language Model Reasoning with Planning Tokens [122.43639723387516]
  Large language models (LLMs) have recently attracted considerable interest for their ability to perform complex reasoning tasks. We propose a hierarchical generation scheme to encourage a more structural generation of chain-of-thought steps. Our approach requires a negligible increase in trainable parameters (0.001%) and can be applied through either full fine-tuning or a more parameter-efficient scheme.
  arXiv  Detail & Related papers  (2023-10-09T13:29:37Z)
• Self-Evaluation Guided Beam Search for Reasoning [61.523627290397556]
  We introduce a stepwise self-evaluation mechanism to guide and calibrate the reasoning process of Large Language Model (LLM) We propose a decoding algorithm integrating the self-evaluation guidance via beam search. Our approach surpasses the corresponding Codex-backboned baselines in few-shot accuracy by $6.34%$, $9.56%$, and $5.46%$ on the GSM8K, AQuA, and StrategyQA.
  arXiv  Detail & Related papers  (2023-05-01T02:37:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.