# ProcBench: Benchmark for Multi-Step Reasoning and Following Procedure

arXiv: https://arxiv.org/abs/2410.03117 · HTML: https://mlsanity.com/paper/2410.03117/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 4 of its 39 references that are in the map.
- First posted on arXiv 4 Oct 2024.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 4 of 4)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [AgentBench: Evaluating LLMs as Agents](https://mlsanity.com/paper/2308.03688/) (2023)
- [MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models](https://mlsanity.com/paper/2309.12284/) (2023)

## Built on by (most cited first, 13 of 13)

- [AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learning](https://mlsanity.com/paper/2505.11896/) (2025)
- [Confidence in Large Language Model Evaluation: A Bayesian Approach to Limited-Sample Challenges](https://mlsanity.com/paper/2504.21303/) (2025)
- [LookAlike: Consistent Distractor Generation in Math MCQs](https://mlsanity.com/paper/2505.01903/) (2025)
- [ProcCtrlBench: Evaluating Process-Level Defects and Control Preservation in LLM Coding Agents](https://mlsanity.com/paper/2605.20251/) (2026)
- [L0-Reasoning Bench: Evaluating Procedural Correctness in Language Models via Simple Program Execution](https://mlsanity.com/paper/2503.22832/) (2025)
- [ConceptMoE: Adaptive Token-to-Concept Compression for Implicit Compute Allocation](https://mlsanity.com/paper/2601.21420/) (2026)
- [MateInfoUB: A Real-World Benchmark for Testing LLMs in Competitive, Multilingual, and Multimodal Educational Tasks](https://mlsanity.com/paper/2507.03162/) (2025)
- [PADME: Procedure Aware DynaMic Execution](https://mlsanity.com/paper/2510.11281/) (2025)
- [Evaluating LLMs' Reasoning Over Ordered Procedural Steps](https://mlsanity.com/paper/2511.04688/) (2025)
- [LLM Swiss Round: Aggregating Multi-Benchmark Performance via Competitive Swiss-System Dynamics](https://mlsanity.com/paper/2512.21010/) (2025)
- [Continual LLM Upcycling: A Predictor-Gated Bank-Wise Sparsity Training Recipe for Dense-to-Sparse LLMs](https://mlsanity.com/paper/2606.10722/) (2026)
- [Constructing Evaluation Datasets for Procedural Reasoning: Balancing Naturalness, Grounding, and Multi-Hop Coverage](https://mlsanity.com/paper/2606.12767/) (2026)
- [One Model, Multiple Compute Budgets: Multi-Ratio ConceptMoE for Flexible Compute Across Turns](https://mlsanity.com/paper/2610.14074/) (2026)

## Abstract (arXiv)

Reasoning is central to a wide range of intellectual activities, and while the capabilities of large language models (LLMs) continue to advance, their performance in reasoning tasks remains limited. The processes and mechanisms underlying reasoning are not yet fully understood, but key elements include path exploration, selection of relevant knowledge, and multi-step inference. Problems are solved through the synthesis of these components. In this paper, we propose a benchmark that focuses on a specific aspect of reasoning ability: the direct evaluation of multi-step inference. To this end, we design a special reasoning task where multi-step inference is specifically focused by largely eliminating path exploration and implicit knowledge utilization. Our dataset comprises pairs of explicit instructions and corresponding questions, where the procedures necessary for solving the questions are entirely detailed within the instructions. This setup allows models to solve problems solely by following the provided directives. By constructing problems that require varying numbers of steps to solve and evaluating responses at each step, we enable a thorough assessment of state-of-the-art LLMs' ability to follow instructions. To ensure the robustness of our evaluation, we include multiple distinct tasks. Furthermore, by comparing accuracy across tasks, utilizing step-aware metrics, and applying separately defined measures of complexity, we conduct experiments that offer insights into the capabilities and limitations of LLMs in reasoning tasks. Our findings have significant implications for the development of LLMs and highlight areas for future research in advancing their reasoning abilities. Our dataset is available at urlhttps://huggingface.co/datasets/ifujisawa/procbench and code at urlhttps://github.com/ifujisawa/proc-bench.