对比单/多智能体在几何题求解中的表现，发现开源模型多智能体更优，闭源模型则视任务而定。
Do Multi-Agents Solve Better Than Single? Evaluating Agentic Frameworks for Diagram-Grounded Geometry Problem Solving and Reasoning
- 用多个智能体分工协作处理几何问题，提升推理效率。
- 开源模型如Qwen-2.5-VL在Geometry3K上最高提分6.8点。
- 闭源模型在新数据集We-Math上多智能体仍有增益，但旧任务优势不明显。
图示几何问题求解是多模态大语言模型（MLLMs）的重要评估基准，但多智能体设计相比单智能体的收益尚不明确。我们系统比较了单智能体与多智能体流程在四个视觉数学基准上的表现：Geometry3K、MathVerse、OlympiadBench和We-Math。对于开源模型，多智能体架构始终带来性能提升。例如，Qwen-2.5-VL (7B) 在Geometry3K上提升6.8分，32B版本提升3.3分；两者在OlympiadBench和We-Math上也进一步增益。相比之下，闭源模型Gemini-2.0-Flash在经典基准上单智能体表现更优，而在较新的We-Math数据集上多智能体仅带来小幅改进。结果表明，多智能体对开源模型有明显优势，且可帮助强大专有系统在新、陌生任务上表现更好，但并非普遍最优。所有代码、数据及推理文件已公开于https://github.com/faiyazabdullah/Interpreter-Solver。
原文摘要 · Abstract (English)
Diagram-grounded geometry problem solving is a critical benchmark for multimodal large language models (MLLMs), yet the benefits of multi-agent design over single-agent remain unclear. We systematically compare single-agent and multi-agent pipelines on four visual math benchmarks: Geometry3K, MathVerse, OlympiadBench, and We-Math. For open-source models, multi-agent consistently improves performance. For example, Qwen-2.5-VL (7B) gains +6.8 points and Qwen-2.5-VL (32B) gains +3.3 on Geometry3K, and both Qwen-2.5-VL variants see further gains on OlympiadBench and We-Math. In contrast, the closed-source Gemini-2.0-Flash generally performs better in single-agent mode on classic benchmarks, while multi-agent yields only modest improvements on the newer We-Math dataset. These findings show that multi-agent pipelines provide clear benefits for open-source models and can assist strong proprietary systems on newer, less familiar benchmarks, but agentic decomposition is not universally optimal. All code, data, and reasoning files are available at https://github.com/faiyazabdullah/Interpreter-Solver
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。