@inproceedings{polyakov-etal-2025-toolreflection,
    title = "{T}ool{R}eflection: Improving Large Language Models for Real-World {API} Calls with Self-Generated Data",
    author = "Polyakov, Gregory  and
      Alimova, Ilseyar  and
      Abulkhanov, Dmitry  and
      Sedykh, Ivan  and
      Bout, Andrey  and
      Nikolenko, Sergey  and
      Piontkovskaya, Irina",
    editor = "Kamalloo, Ehsan  and
      Gontier, Nicolas  and
      Lu, Xing Han  and
      Dziri, Nouha  and
      Murty, Shikhar  and
      Lacoste, Alexandre",
    booktitle = "Proceedings of the 1st Workshop for Research on Agent Language Models (REALM 2025)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/display_plenaries/2025.realm-1.14/",
    pages = "184--199",
    ISBN = "979-8-89176-264-0",
    abstract = "While open-source large language models (LLMs) have advanced in leveraging third-party tools, significant challenges remain in real-world API usage, where behavior is unpredictable or poorly specified. Existing benchmarks often fail to capture this complexity. We propose ToolReflection, a novel method that improves LLMs' ability to self-correct API calls by utilizing real-time API feedback. We also introduce new datasets specifically designed to test model performance under realistic conditions. In ToolReflection, models undergo instruction tuning on a dataset augmented with self-generated errors and corrections. Our evaluation across ToolAlpaca, ToolBench benchmarks, and three newly developed datasets (GPT4Tools-OOD, GPT4Tools-OOD-Hard, and Multistep-100) demonstrates its effectiveness. ToolReflection boosts overall success rates by 25.4{\%} on GPT4Tools-OOD, 56.2{\%} on GPT4Tools-OOD-Hard, and 4{\%} on Multistep-100, outperforming original models. On ToolAlpaca, we show a 14{\%} improvement in the ``Simulated'' setting and 10.5{\%} in the ``Real-world'' scenario. Our error analysis highlights ToolReflection significantly enhances recovery from incorrect tool calls, even with incomplete or erroneous API documentation. We have released the code, prompts, and data at https://github.com/polgrisha/ToolReflection."
}