Fin-PRM: A Domain-Specialized Process Reward Model for Financial Reasoning in Large Language Models

TL;DR

Fin-PRM is a specialized process reward model designed for financial reasoning in large language models, improving intermediate step verification and overall reasoning accuracy.

Contribution

It introduces a domain-specific, trajectory-aware reward model trained on a high-quality financial reasoning dataset, enhancing reasoning tasks in finance.

Findings

Fin-PRM outperforms general PRMs on financial benchmarks.

It improves offline trajectory selection and test-time inference.

The model effectively integrates step- and trajectory-level rewards.

Abstract

Process Reward Models (PRMs) supervise intermediate reasoning steps in large language models (LLMs), but existing PRMs are mainly trained on general-domain data and struggle with the structured, symbolic, and fact-sensitive nature of financial reasoning. Financial tasks require not only correct final answers but also verifiable intermediate steps grounded in domain knowledge. In this paper, we propose Fin-PRM, a domain-specialized, trajectory-aware PRM for financial reasoning that jointly models step-level correctness and trajectory-level coherence, producing binary supervision signals for both local and global reasoning quality. To support reliable supervision, we construct a high-quality financial reasoning dataset of 3K trajectories, where step- and trajectory-level labels are automatically derived from multi-source reward signals, including Monte Carlo rollouts, LLM-based evaluation, and explicit financial knowledge verification. Fin-PRM defines a unified ranking score that integrates step- and trajectory-level rewards, enabling consistent use across multiple settings. We evaluate Fin-PRM in three scenarios: (1) offline trajectory selection for supervised fine-tuning, (2) reward-guided Best-of-$N$ inference for test-time scaling, and (3) process-aware reward shaping for reinforcement learning. Experiments on financial reasoning benchmarks, including CFLUE and FinQA, show that Fin-PRM consistently outperforms general-purpose PRMs and strong baselines. Our project resources will be available at https://github.com/aliyun/qwen-dianjin.