loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Paper

Authors: Miloš Košprdić 1 ; Adela Ljajić 1 ; Darija Medvecki 1 ; Bojana Bašaragin 1 and Nikola Milošević 2 ; 1

Affiliations: 1 The Institute for Artificial Intelligence Research and Development of Serbia, Fruškogorska 1, Novi Sad, Republic of Serbia ; 2 Bayer A.G. Reaserch and Development, Mullerstrasse 173, Berlin, Germany

Keyword(s): Claim Verification, Deep Learning Models, Natural Language Inference, PubMed, SciFact Dataset.

Abstract: This paper introduces the foundation for the third component of a pioneering open-source scientific question-answering system. The system is designed to provide referenced, automatically vetted, and verifiable answers in the scientific domain where hallucinations and misinformation are intolerable. This Verification Engine is based on models fine-tuned for the Natural Language Inference task using an additionally processed SciFact dataset. Our experiments, involving eight fine-tuned models based on RoBERTa Large, XLM RoBERTa Large, DeBERTa, and DeBERTa SQuAD, show promising results. Notably, the DeBERTa model fine-tuned on our dataset achieved the highest F1 score of 88%. Furthermore, evaluating our best model on the HealthVer dataset resulted in an F1 score of 48%, outperforming other models by more than 12%. Additionally, our model demonstrated superior performance with a 7% absolute increase in F1 score compared to the best-performing GPT-4 model on the same test set in a zero-sho t regime. These findings suggest that our system can significantly enhance scientists’ productivity while fostering trust in the use of generative language models in scientific environments. (More)

CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 3.147.73.41

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Košprdić, M., Ljajić, A., Medvecki, D., Bašaragin, B. and Milošević, N. (2024). Scientific Claim Verification with Fine-Tuned NLI Models. In Proceedings of the 16th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KMIS; ISBN 978-989-758-716-0; ISSN 2184-3228, SciTePress, pages 15-25. DOI: 10.5220/0012900000003838

@conference{kmis24,
author={Miloš Košprdić and Adela Ljajić and Darija Medvecki and Bojana Bašaragin and Nikola Milošević},
title={Scientific Claim Verification with Fine-Tuned NLI Models},
booktitle={Proceedings of the 16th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KMIS},
year={2024},
pages={15-25},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012900000003838},
isbn={978-989-758-716-0},
issn={2184-3228},
}

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - KMIS
TI - Scientific Claim Verification with Fine-Tuned NLI Models
SN - 978-989-758-716-0
IS - 2184-3228
AU - Košprdić, M.
AU - Ljajić, A.
AU - Medvecki, D.
AU - Bašaragin, B.
AU - Milošević, N.
PY - 2024
SP - 15
EP - 25
DO - 10.5220/0012900000003838
PB - SciTePress