Graph of Verification: Structured Verification of LLM Reasoning with Directed Acyclic Graphs

Verifying the reliability of complex, multi-step reasoning in Large Language Models (LLMs) remains a fundamental challenge, as existing methods often lack both faithfulness and precision. To address this issue, we propose the Graph of Verification (GoV) framework. GoV offers three key contributions: First, it explicitly models the underlying deductive process as a directed acyclic graph (DAG), whether this structure is implicit or explicitly constructed. Second, it enforces a topological order over the DAG to guide stepwise verification. Third, GoV introduces the notion of customizable node blocks, which flexibly define the verification granularity, from atomic propositions to full paragraphs, while ensuring that all requisite premises derived from the graph are provided as contextual input for each verification unit. We evaluate GoV on the Number Triangle Summation task and the ProcessBench benchmark with varying levels of reasoning complexity. Experimental results show that GoV substantially improves verification accuracy, faithfulness, and error localization when compared to conventional end-to-end verification approaches. Our code and data are available atthis https URL.

@article{fang2025_2506.12509,
  title={ Graph of Verification: Structured Verification of LLM Reasoning with Directed Acyclic Graphs },
  author={ Jiwei Fang and Bin Zhang and Changwei Wang and Jin Wan and Zhiwei Xu },
  journal={arXiv preprint arXiv:2506.12509},
  year={ 2025 }
}