让大模型推理结果更可靠，自动选工具验证。
VerifiAgent: a Unified Verification Agent in Language Model Reasoning
- 分层验证：先检查回答完整一致，再按类型选数学/逻辑/常识工具
- 在各类推理任务中表现优于传统验证方法，提升准确率
- 适合需要高可信度推理的场景，如数学证明、复杂决策
大语言模型虽具强大推理能力，但常产生不可靠或错误回应。现有验证方法多局限于特定模型或领域，计算开销大且难以跨任务扩展。为此，我们提出VerifiAgent，一种统一验证代理，集成双层验证机制：元验证评估模型输出的完整性与一致性；基于任务类型的自适应工具验证，可自主选择数学、逻辑或常识类验证工具。该方法在不同验证场景下兼具高效性与鲁棒性。实验表明，VerifiAgent在所有推理任务中均优于基准验证方法（如演绎验证器、反向验证器）。同时，可通过验证反馈进一步提升推理准确性。在推理扩容中，相比现有过程奖励模型，可用更少生成样本和更低成本获得更优数学推理结果。代码已开源：https://github.com/Jiuzhouh/VerifiAgent。
原文摘要 · Abstract (English)
Large language models demonstrate remarkable reasoning capabilities but often produce unreliable or incorrect responses. Existing verification methods are typically model-specific or domain-restricted, requiring significant computational resources and lacking scalability across diverse reasoning tasks. To address these limitations, we propose VerifiAgent, a unified verification agent that integrates two levels of verification: meta-verification, which assesses completeness and consistency in model responses, and tool-based adaptive verification, where VerifiAgent autonomously selects appropriate verification tools based on the reasoning type, including mathematical, logical, or commonsense reasoning. This adaptive approach ensures both efficiency and robustness across different verification scenarios. Experimental results show that VerifiAgent outperforms baseline verification methods (e.g., deductive verifier, backward verifier) among all reasoning tasks. Additionally, it can further enhance reasoning accuracy by leveraging feedback from verification results. VerifiAgent can also be effectively applied to inference scaling, achieving better results with fewer generated samples and costs compared to existing process reward models in the mathematical reasoning domain. Code is available at https://github.com/Jiuzhouh/VerifiAgent
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。