Skoltech at TextGraphs-17 Shared Task: Finding GPT-4 Prompting Strategies for Multiple Choice Questions

Maria Lysyuk, Pavel Braslavski

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In this paper, we present our solution to the TextGraphs-17 Shared Task on Text-Graph Representations for Knowledge Graph Question Answering (KGQA). GPT-4 alone, with chain-of-thought reasoning and a given set of answers, achieves an F1 score of 0.78. By employing subgraph size as a feature, Wikidata answer description as an additional context, and a question rephrasing technique, we further strengthen this result. These tricks help to answer questions that were not initially answered and to eliminate irrelevant, identical answers. We have managed to achieve an F1 score of 0.83 and took 2nd place, improving the score by 0.05 over the baseline. An open implementation of our method is available on GitHub.

Original languageEnglish
Title of host publicationTextGraphs at ACL 2024 - Proceedings of TextGraphs-17
Subtitle of host publicationGraph-Based Methods for Natural Language Processing, 62nd Annual Meeting of the Association of Computational Linguistics
EditorsDmitry Ustalov, Yanjun Gao, Alexander Panchenko, Elena Tutubalina, Irina Nikishina, Arti Ramesh, Andrey Sakhovskiy, Ricardo Usbeck, Gerald Penn, Marco Valentino
PublisherAssociation for Computational Linguistics (ACL)
Pages149-153
Number of pages5
ISBN (Electronic)9798891761452
Publication statusPublished - 2024
Event17th Graph-Based Methods for Natural Language Processing, TextGraphs 2024 - co-located with the 62nd Annual Meeting of the Association of Computational Linguistics, ACL 2024 - Bangkok, Thailand
Duration: Aug 15 2024 → …

Publication series

NameTextGraphs at ACL 2024 - Proceedings of TextGraphs-17: Graph-Based Methods for Natural Language Processing, 62nd Annual Meeting of the Association of Computational Linguistics

Conference

Conference17th Graph-Based Methods for Natural Language Processing, TextGraphs 2024 - co-located with the 62nd Annual Meeting of the Association of Computational Linguistics, ACL 2024
Country/TerritoryThailand
CityBangkok
Period8/15/24 → …

ASJC Scopus subject areas

  • Language and Linguistics
  • Computational Theory and Mathematics
  • Software
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'Skoltech at TextGraphs-17 Shared Task: Finding GPT-4 Prompting Strategies for Multiple Choice Questions'. Together they form a unique fingerprint.

Cite this