EMNLP 2024industry21 citations

Granite-Function Calling Model: Introducing Function Calling Abilities via Multi-task Learning of Granular Tasks

Ibrahim Abdelaziz, Kinjal Basu, Mayank Agarwal, Sadhana Kumaravel, Matthew Stallone, Rameswar Panda, Yara Rizk, G P Shrivatsa Bhargav

Abstract

An emergent research trend explores the use of Large Language Models (LLMs) as the backbone of agentic systems (e.g., SWE-Bench, Agent-Bench). To fulfill LLMs’ potential as autonomous agents, they must be able to identify, call, and interact with a variety of external tools and application program interfaces (APIs). This capability of LLMs, commonly termed function calling, leads to a myriad of advantages such as access to current and domain-specific information in databases and the outsourcing of tasks that can be reliably performed by tools. In this work, we introduce Granite-20B-FunctionCalling, a model trained using a multi-task training approach on seven fundamental tasks encompassed in function calling. Our comprehensive evaluation on multiple out-of-domain datasets, which compares Granite-20B-FunctionCalling to more than 15 other best proprietary and open models, shows that Granite-20B-FunctionCalling has better generalizability on multiple tasks across seven different evaluation benchmarks. Moreover, Granite-20B-FunctionCalling shows the best performance among all open models and ranks among the top on the Berkeley Function Calling Leaderboard (BFCL).

BibTeX
@inproceedings{abdelaziz-etal-2024-granite,
    title = "Granite-Function Calling Model: Introducing Function Calling Abilities via Multi-task Learning of Granular Tasks",
    author = "Abdelaziz, Ibrahim  and
      Basu, Kinjal  and
      Agarwal, Mayank  and
      Kumaravel, Sadhana  and
      Stallone, Matthew  and
      Panda, Rameswar  and
      Rizk, Yara  and
      Bhargav, G P Shrivatsa  and
      Crouse, Maxwell  and
      Gunasekara, Chulaka  and
      Ikbal, Shajith  and
      Joshi, Sachindra  and
      Karanam, Hima  and
      Kumar, Vineet  and
      Munawar, Asim  and
      Neelam, Sumit  and
      Raghu, Dinesh  and
      Sharma, Udit  and
      Soria, Adriana Meza  and
      Sreedhar, Dheeraj  and
      Venkateswaran, Praveen  and
      Unuvar, Merve  and
      Cox, David Daniel  and
      Roukos, Salim  and
      Lastras, Luis A.  and
      Kapanipathi, Pavan",
    editor = "Dernoncourt, Franck  and
      Preo{\c{t}}iuc-Pietro, Daniel  and
      Shimorina, Anastasia",
    booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track",
    month = nov,
    year = "2024",
    address = "Miami, Florida, US",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.emnlp-industry.85/",
    doi = "10.18653/v1/2024.emnlp-industry.85",
    pages = "1131--1139"
}
Granite-Function Calling Model: Introducing Function Calling Abilities via Multi-task Learning of Granular Tasks · EMNLP 2024