Search papers, labs, and topics across Lattice.
EvoAgent is introduced, an LLM agent framework that evolves skills through user feedback and a multi-agent delegation mechanism. Skills are represented as multi-file capability units with triggering mechanisms, enabling continuous generation and optimization. Experiments in foreign trade scenarios show that GPT5.2 integrated with EvoAgent improves significantly in professionalism, accuracy, and utility, increasing the average evaluation score by 28%.
LLM agent performance hinges as much on the agent architecture's synergy with the model as on the model's intrinsic capabilities, challenging the assumption that bigger models automatically translate to better agents.
This paper proposes EvoAgent - an evolvable large language model (LLM) agent framework that integrates structured skill learning with a hierarchical sub-agent delegation mechanism. EvoAgent models skills as multi-file structured capability units equipped with triggering mechanisms and evolutionary metadata, and enables continuous skill generation and optimization through a user-feedback-driven closed-loop process. In addition, by incorporating a three-stage skill matching strategy and a three-layer memory architecture, the framework supports dynamic task decomposition for complex problems and long-term capability accumulation. Experimental results based on real-world foreign trade scenarios demonstrate that, after integrating EvoAgent, GPT5.2 achieves significant improvements in professionalism, accuracy, and practical utility. Under a five-dimensional LLM-as-Judge evaluation protocol, the overall average score increases by approximately 28%. Further model transfer experiments indicate that the performance of an agent system depends not only on the intrinsic capabilities of the underlying model, but also on the degree of synergy between the model and the agent architecture.
