Leveraging small language models for Text2SPARQLtasks to improve the resilience of AI assistance
F. Brei, J. Frey, and L. Meyer. Proceedings of the Third International Workshop on Linked Data-driven Resilience Research 2024 (D2R2'24), colocated with ESWC 2024, volume 3707 of CEUR-WS, (2024)
DOI: 10.48550/arXiv.2405.17076
Abstract
In this work we will show that language models with less than one billion parameters can be used to translate natural language to SPARQL queries after fine-tuning. Using three different datasets ranging from academic to real world, we identify prerequisites that the training data must fulfill in order for the training to be successful. The goal is to empower users of semantic web technology to use AI assistance with affordable commodity hardware, making them more resilient against external factors
%0 Conference Paper
%1 Brei2024Leveragingsmalllanguage
%A Brei, Felix
%A Frey, Johannes
%A Meyer, Lars-Peter
%B Proceedings of the Third International Workshop on Linked Data-driven Resilience Research 2024 (D2R2'24), colocated with ESWC 2024
%D 2024
%E Holze, Julia
%E Tramp, Sebastian
%E Martin, Michael
%E Auer, Sören
%E Usbeck, Ricardo
%E Krdzavac, Nenad
%K es frey group_aksw lpmeyer
%R 10.48550/arXiv.2405.17076
%T Leveraging small language models for Text2SPARQLtasks to improve the resilience of AI assistance
%U https://ceur-ws.org/Vol-3707/D2R224_paper_5.pdf
%V 3707
%X In this work we will show that language models with less than one billion parameters can be used to translate natural language to SPARQL queries after fine-tuning. Using three different datasets ranging from academic to real world, we identify prerequisites that the training data must fulfill in order for the training to be successful. The goal is to empower users of semantic web technology to use AI assistance with affordable commodity hardware, making them more resilient against external factors
@inproceedings{Brei2024Leveragingsmalllanguage,
abstract = {In this work we will show that language models with less than one billion parameters can be used to translate natural language to SPARQL queries after fine-tuning. Using three different datasets ranging from academic to real world, we identify prerequisites that the training data must fulfill in order for the training to be successful. The goal is to empower users of semantic web technology to use AI assistance with affordable commodity hardware, making them more resilient against external factors},
added-at = {2024-06-18T09:46:40.000+0200},
author = {Brei, Felix and Frey, Johannes and Meyer, Lars-Peter},
biburl = {https://www.bibsonomy.org/bibtex/2714e885318c7b292ec459439f12498e1/aksw},
booktitle = {Proceedings of the Third International Workshop on Linked Data-driven Resilience Research 2024 (D2R2'24), colocated with ESWC 2024},
doi = {10.48550/arXiv.2405.17076},
editor = {Holze, Julia and Tramp, Sebastian and Martin, Michael and Auer, Sören and Usbeck, Ricardo and Krdzavac, Nenad},
interhash = {6177dc6e8f694511545f88f123e9056e},
intrahash = {714e885318c7b292ec459439f12498e1},
keywords = {es frey group_aksw lpmeyer},
series = {CEUR-WS},
timestamp = {2024-06-18T09:46:40.000+0200},
title = {Leveraging small language models for Text2SPARQLtasks to improve the resilience of AI assistance},
url = {https://ceur-ws.org/Vol-3707/D2R224_paper_5.pdf},
volume = 3707,
year = 2024
}