EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LettRAGraph at LLMs4OL 2026 Task Flagship: Prompt-Guided Ontology Extraction With Ensembling

Pasquale Lisena, Julien Plu, Oscar Moreno Escobar, Edouard Trouillez, Raphael Troncy

Open Conference Proceedings · 03.11.2026 · 11

DOI: 10.52825/ocp.v11i.3655 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Open Conference Proceedings, online erschienen am 03.11.2026. Verfasst von Pasquale Lisena und 4 weiteren Autor:innen. Die Arbeit nennt 13 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Bpifrance.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present a system for constructing ontology graphs from text using open-weight large language models. Our approach combines structured prompting, predicate normalization, and structural filtering. Prompt ablation experiments show that explicit taxonomy rules and closed-world constraints improve graph quality. We also introduce a conservative ensemble that retains only triples shared by two prompt configurations. On the official test set, the selected ensemble achieves Graph Similarity scores of 0.3584, 0.4872, and 0.5162 under exact, fuzzy, and semantic matching.

Förderung

  • Bpifrance

Zitieren

APA

Lisena, P., Plu, J., Escobar, O. M., Trouillez, E., & Troncy, R. (2026). LettRAGraph at LLMs4OL 2026 Task Flagship: Prompt-Guided Ontology Extraction With Ensembling. Open Conference Proceedings, 11. https://doi.org/10.52825/ocp.v11i.3655

Vancouver

Lisena P, Plu J, Escobar OM, Trouillez E, Troncy R. LettRAGraph at LLMs4OL 2026 Task Flagship: Prompt-Guided Ontology Extraction With Ensembling. Open Conference Proceedings. 2026;11. doi:10.52825/ocp.v11i.3655

BibTeX
@article{Lisena2026,
  author = {Lisena, Pasquale and Plu, Julien and Escobar, Oscar Moreno and Trouillez, Edouard and Troncy, Raphael},
  title = {LettRAGraph at LLMs4OL 2026 Task Flagship: Prompt-Guided Ontology Extraction With Ensembling},
  journal = {Open Conference Proceedings},
  year = {2026},
  volume = {11},
  doi = {10.52825/ocp.v11i.3655}
}
Kostenlos starten →