2024/11/02 by Bin Lei, Yuchen Li, Lei, Bin +17 · 4 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Multi-Agent Systems and Negotiation
paper · pdf · doi:10.48550/arxiv.2411.01114
openalex publication_date 2024/11/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Despite the impressive capabilities of large language models (LLMs), they currently exhibit two primary limitations, \textbf\uppercase\expandafter\romannumeral 1: They struggle to autonomously solve the real world engineering problem. \textbf\uppercase\expandafter\romannumeral 2: They remain challenged in reasoning through complex logic problems. To address these challenges, we developed the 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 Infant Agent, GPT-4o's accuracy on the SWE-bench-lite dataset rises from 0.33% to 30%, and in the AIME-2024 mathematics competition, it increases GPT-4o's accuracy from 13.3% to 37%.