eidos: A modular approach to external function integration in LLMs

Function calling allows Large Language Models (LLMs) to execute a wide range of tasks, from data analysis and mathematical computations to interacting with web services and other software systems. By harnessing the power of external tooling, LLMs can provide more dynamic, context-aware responses. Ho...

Full description

Saved in:
Bibliographic Details
Main Authors: José F. Aldana-Martín, Antonio Benítez-Hidalgo, José F. Aldana-Montes
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:SoftwareX
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2352711025002560
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849236513293860864
author José F. Aldana-Martín
Antonio Benítez-Hidalgo
José F. Aldana-Montes
author_facet José F. Aldana-Martín
Antonio Benítez-Hidalgo
José F. Aldana-Montes
author_sort José F. Aldana-Martín
collection DOAJ
description Function calling allows Large Language Models (LLMs) to execute a wide range of tasks, from data analysis and mathematical computations to interacting with web services and other software systems. By harnessing the power of external tooling, LLMs can provide more dynamic, context-aware responses. However, errors in the model’s understanding of the request can lead to misinterpretations of the intended actions, resulting in function calls that are either irrelevant or incorrect for the task at hand. Without proper validation and control mechanisms, the parameters expected by the function may not align with those provided by the model, leading to incorrect operations or failures in task execution. In this paper, we present eidos, a software tool designed to streamline the integration of functions within LLMs. eidos acts as an intermediary, enabling both the seamless execution and validation of functions by LLMs. By leveraging its modular architecture, function definitions can be injected into the LLM context and invoked as if they were native functions via an API.
format Article
id doaj-art-4bca3d10a1b048c0b8d95323aa6e7268
institution Kabale University
issn 2352-7110
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series SoftwareX
spelling doaj-art-4bca3d10a1b048c0b8d95323aa6e72682025-08-20T04:02:13ZengElsevierSoftwareX2352-71102025-09-013110229010.1016/j.softx.2025.102290eidos: A modular approach to external function integration in LLMsJosé F. Aldana-Martín0Antonio Benítez-Hidalgo1José F. Aldana-Montes2Corresponding author at: ITIS Software, Edificio de Investigación Ada Byron, University of Málaga, Málaga, 29071, Spain.; ITIS Software, Edificio de Investigación Ada Byron, University of Málaga, Málaga, 29071, Spain; Dept. de Lenguajes y Ciencias de la Computación, University of Málaga, Málaga, 29071, SpainITIS Software, Edificio de Investigación Ada Byron, University of Málaga, Málaga, 29071, Spain; Dept. de Lenguajes y Ciencias de la Computación, University of Málaga, Málaga, 29071, SpainITIS Software, Edificio de Investigación Ada Byron, University of Málaga, Málaga, 29071, Spain; Dept. de Lenguajes y Ciencias de la Computación, University of Málaga, Málaga, 29071, SpainFunction calling allows Large Language Models (LLMs) to execute a wide range of tasks, from data analysis and mathematical computations to interacting with web services and other software systems. By harnessing the power of external tooling, LLMs can provide more dynamic, context-aware responses. However, errors in the model’s understanding of the request can lead to misinterpretations of the intended actions, resulting in function calls that are either irrelevant or incorrect for the task at hand. Without proper validation and control mechanisms, the parameters expected by the function may not align with those provided by the model, leading to incorrect operations or failures in task execution. In this paper, we present eidos, a software tool designed to streamline the integration of functions within LLMs. eidos acts as an intermediary, enabling both the seamless execution and validation of functions by LLMs. By leveraging its modular architecture, function definitions can be injected into the LLM context and invoked as if they were native functions via an API.http://www.sciencedirect.com/science/article/pii/S2352711025002560Large Language ModelsFunction callingValidationRetrieval-Augmented Generation
spellingShingle José F. Aldana-Martín
Antonio Benítez-Hidalgo
José F. Aldana-Montes
eidos: A modular approach to external function integration in LLMs
SoftwareX
Large Language Models
Function calling
Validation
Retrieval-Augmented Generation
title eidos: A modular approach to external function integration in LLMs
title_full eidos: A modular approach to external function integration in LLMs
title_fullStr eidos: A modular approach to external function integration in LLMs
title_full_unstemmed eidos: A modular approach to external function integration in LLMs
title_short eidos: A modular approach to external function integration in LLMs
title_sort eidos a modular approach to external function integration in llms
topic Large Language Models
Function calling
Validation
Retrieval-Augmented Generation
url http://www.sciencedirect.com/science/article/pii/S2352711025002560
work_keys_str_mv AT josefaldanamartin eidosamodularapproachtoexternalfunctionintegrationinllms
AT antoniobenitezhidalgo eidosamodularapproachtoexternalfunctionintegrationinllms
AT josefaldanamontes eidosamodularapproachtoexternalfunctionintegrationinllms