让大模型更智能地解决复杂工程与逻辑问题，同时大幅降低使用成本。
Infant Agent: A Tool-Integrated, Logic-Driven Agent with Cost-Effective API Usage
- 整合工具与分层管理，增强大模型持续推理能力
- 在SWE-bench-lite上准确率从0.33%提升至30%
- 适合需要高效推理和低成本部署的工程应用
尽管大型语言模型（LLMs）表现出色，但仍存在两大局限：难以自主解决真实世界工程问题，且在复杂逻辑推理上表现不足。为此，我们开发了Infant Agent，集成任务感知功能、操作符、分层管理系统和记忆检索机制。这些组件共同使大模型能够进行长时间推理，高效处理多步骤复杂任务，同时显著降低API调用成本。使用Infant Agent后，GPT-4o在SWE-bench-lite数据集上的准确率从0.33%提升至30%，在AIME-2024数学竞赛中从13.3%提升至37%。
原文摘要 · Abstract (English)
Despite the impressive capabilities of large language models (LLMs), they currently exhibit two primary limitations, \textbf{\uppercase\expandafter{\romannumeral 1}}: They struggle to \textbf{autonomously solve the real world engineering problem}. \textbf{\uppercase\expandafter{\romannumeral 2}}: They remain \textbf{challenged in reasoning through complex logic problems}. To address these challenges, we developed the \textsc{Infant Agent}, integrating task-aware functions, operators, a hierarchical management system, and a memory retrieval mechanism. Together, these components enable large language models to sustain extended reasoning processes and handle complex, multi-step tasks efficiently, all while significantly reducing API costs. Using the \textsc{Infant Agent}, GPT-4o's accuracy on the SWE-bench-lite dataset rises from $\mathbf{0.33\%}$ to $\mathbf{30\%}$, and in the AIME-2024 mathematics competition, it increases GPT-4o's accuracy from $\mathbf{13.3\%}$ to $\mathbf{37\%}$.
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。