2026-07-23T07:36:49Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:18515 2024-03-06T10:32:08Z ddc:004 open_access

Question Answering over Linked Data with GPT-3 Faria, Bruno Perdigão, Dylan Gonçalo Oliveira, Hugo SPARQL Generation Prompt Engineering Few-Shot Learning Question Answering GPT-3 This paper explores GPT-3 for answering natural language questions over Linked Data. Different engines of the model and different approaches are adopted for answering questions in the QALD-9 dataset, namely: zero and few-shot SPARQL generation, as well as fine-tuning in the training portion of the dataset. Answers retrieved by the generated queries and answers generated directly by the model are also compared. Overall results are generally poor, but several insights are provided on using GPT-3 for the proposed task. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Bruno Faria and Dylan Perdigão and Hugo Gonçalo Oliveira 2023 Is Part Of OASIcs, Volume 113, 12th Symposium on Languages, Applications and Technologies (SLATE 2023) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/OASIcs.SLATE.2023.1 urn:nbn:de:0030-drops-185155 https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.SLATE.2023.1 eng https://creativecommons.org/licenses/by/4.0/legalcode