ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents

• URL: http://arxiv.org/abs/2601.12294v1
• Date: Sun, 18 Jan 2026 07:48:36 GMT
• Title: ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents
• Authors: Dawei Li, Yuguang Yao, Zhen Tan, Huan Liu, Ruocheng Guo,
• Abstract summary: Reward-guided search methods have demonstrated strong potential in enhancing tool-using agents.<n>There is a lack of systematic and reliable evaluation benchmarks for PRMs in tool-using settings.<n>We introduce ToolPRMBench, a large-scale benchmark specifically designed to evaluate PRMs for tool-using agents.
• Score: 31.77712252239516
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reward-guided search methods have demonstrated strong potential in enhancing tool-using agents by effectively guiding sampling and exploration over complex action spaces. As a core design, those search methods utilize process reward models (PRMs) to provide step-level rewards, enabling more fine-grained monitoring. However, there is a lack of systematic and reliable evaluation benchmarks for PRMs in tool-using settings. In this paper, we introduce ToolPRMBench, a large-scale benchmark specifically designed to evaluate PRMs for tool-using agents. ToolPRMBench is built on top of several representative tool-using benchmarks and converts agent trajectories into step-level test cases. Each case contains the interaction history, a correct action, a plausible but incorrect alternative, and relevant tool metadata. We respectively utilize offline sampling to isolate local single-step errors and online sampling to capture realistic multi-step failures from full agent rollouts. A multi-LLM verification pipeline is proposed to reduce label noise and ensure data quality. We conduct extensive experiments across large language models, general PRMs, and tool-specialized PRMs on ToolPRMBench. The results reveal clear differences in PRM effectiveness and highlight the potential of specialized PRMs for tool-using. Code and data will be released at https://github.com/David-Li0406/ToolPRMBench.

Related papers

• MCPAgentBench: A Real-world Task Benchmark for Evaluating LLM Agent MCP Tool Use [12.220519951554133]
  MCPAgentBench is a benchmark based on real-world MCP definitions to evaluate the tool-use capabilities of agents.<n>The evaluation employs a dynamic sandbox environment that presents agents with candidate tool lists containing distractors.<n>Experiments conducted on various latest mainstream Large Language Models reveal significant performance differences in handling complex, multi-step tool invocations.
  arXiv  Detail & Related papers  (2025-12-31T02:09:48Z)
• ML-Tool-Bench: Tool-Augmented Planning for ML Tasks [23.54937738755734]
  We introduce a benchmark for evaluating tool-augmented machine learning agents.<n>Our benchmark goes beyond traditional tool-use evaluation by incorporating an in-memory named object management.<n>Our approach improves over ReAct by 16.52 percentile positions, taking the median across all Kaggle challenges.
  arXiv  Detail & Related papers  (2025-11-29T23:59:40Z)
• PRInTS: Reward Modeling for Long-Horizon Information Seeking [74.14496236655911]
  We introduce PRInTS, a generative PRM trained with dual capabilities.<n>We show that PRInTS enhances information-seeking abilities of open-source models as well as specialized agents.
  arXiv  Detail & Related papers  (2025-11-24T17:09:43Z)
• AgentPRM: Process Reward Models for LLM Agents via Step-Wise Promise and Progress [71.02263260394261]
  Large language models (LLMs) still encounter challenges in multi-turn decision-making tasks.<n>We build process reward models (PRMs) to evaluate each decision and guide the agent's decision-making process.<n>AgentPRM captures both the interdependence between sequential decisions and their contribution to the final goal.
  arXiv  Detail & Related papers  (2025-11-11T14:57:54Z)
• One Model to Critique Them All: Rewarding Agentic Tool-Use via Efficient Reasoning [54.580646706013965]
  Reward models (RMs) play a critical role in aligning large language models with human preferences.<n>We introduce ToolRM, a family of lightweight generative RMs tailored for general tool-use scenarios.<n>To build these models, we propose a novel pipeline that constructs pairwise preference data using rule-based scoring and multidimensional sampling.
  arXiv  Detail & Related papers  (2025-10-30T06:08:27Z)
• TRAJECT-Bench:A Trajectory-Aware Benchmark for Evaluating Agentic Tool Use [74.47746287181383]
  Large language model (LLM)-based agents increasingly rely on tool use to complete real-world tasks.<n>We introduce TRAJECT-Bench, a trajectory-aware benchmark to comprehensively evaluate LLMs' tool use capability.
  arXiv  Detail & Related papers  (2025-10-06T07:30:25Z)
• LaMDAgent: An Autonomous Framework for Post-Training Pipeline Optimization via LLM Agents [3.6117068575553595]
  We introduce LaMDAgent, a framework that autonomously constructs and optimize full post-training pipelines.<n>LaMDAgent improves tool-use accuracy by 9.0 points while preserving instruction-following capabilities.<n>It uncovers effective post-training strategies that are often overlooked by conventional human-driven exploration.
  arXiv  Detail & Related papers  (2025-05-28T04:30:51Z)
• T^2Agent A Tool-augmented Multimodal Misinformation Detection Agent with Monte Carlo Tree Search [51.91311158085973]
  multimodal misinformation often arises from mixed forgery sources, requiring dynamic reasoning and adaptive verification.<n>We propose T2Agent, a novel misinformation detection agent that incorporates a toolkit with Monte Carlo Tree Search.<n>Extensive experiments show that T2Agent consistently outperforms existing baselines on challenging mixed-source multimodal misinformation benchmarks.
  arXiv  Detail & Related papers  (2025-05-26T09:50:55Z)
• Iterative Tool Usage Exploration for Multimodal Agents via Step-wise Preference Tuning [68.00304954972232]
  Multimodal agents, which integrate a controller e.g., a vision language model, with external tools, have demonstrated remarkable capabilities in tackling complex multimodal tasks.<n>Existing approaches for training these agents depend on extensive human-annotated task-answer pairs and tool trajectories.<n>We propose an iterative tool usage exploration method for multimodal agents without any pre-collected data, namely SPORT.<n>SPORT has four iterative components: task synthesis, step sampling, step verification, and preference tuning.
  arXiv  Detail & Related papers  (2025-04-30T12:01:27Z)

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.