@inproceedings{hao-etal-2026-recreate,
    title = "{R}e{C}reate: Reasoning and Creating Domain Agents Driven by Experience",
    author = "Hao, Zhezheng  and
      Wang, Hong  and
      Luo, Jian  and
      Zhang, Jianqing  and
      Zhou, Yuyan  and
      Lin, Qiang  and
      Wang, Can  and
      Dong, Hande  and
      Chen, Jiawei",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1432/",
    pages = "31018--31046",
    ISBN = "979-8-89176-390-6",
    abstract = "Large Language Model (LLM) agents are reshaping the industrial landscape. However, most practical agents remain human-designed because tasks differ widely, making them labor-intensive to build. This situation poses a central question: \textit{can we automatically create and adapt domain agents in the wild?} While several recent approaches have sought to automate agent creation, they typically treat agent generation as a black-box procedure and rely solely on final performance metrics to guide the process. Such strategies overlook critical evidence explaining why an agent succeeds or fails, and often require high computational costs. To address these limitations, we propose \textit{ReCreate}, an experience-driven framework for the automatic creation of domain agents. ReCreate systematically leverages agent interaction histories, which provide rich concrete signals on both the causes of success or failure and the avenues for improvement. Specifically, we introduce an \textit{agent-as-optimizer} paradigm that effectively learns from experience via three key components: (i) an experience storage and retrieval mechanism for on-demand inspection; (ii) a reasoning{--}creating synergy pipeline that map execution experience into scaffold edits; and (iii) hierarchical updates that abstract instance-level details into reusable domain patterns. In experiments across diverse domains, ReCreate consistently outperforms human-designed agents and existing automated agent generation methods, even when starting from minimal seed scaffolds."
}