COLING 2024main2 citations

How to Encode Domain Information in Relation Classification

Elisa Bassignana, Viggo Unmack Gascou, Frida Nøhr Laustsen, Gustav Kristensen, Marie Haahr Petersen, Rob van der Goot, Barbara Plank

Abstract

Current language models require a lot of training data to obtain high performance. For Relation Classification (RC), many datasets are domain-specific, so combining datasets to obtain better performance is non-trivial. We explore a multi-domain training setup for RC, and attempt to improve performance by encoding domain information. Our proposed models improve > 2 Macro-F1 against the baseline setup, and our analysis reveals that not all the labels benefit the same: The classes which occupy a similar space across domains (i.e., their interpretation is close across them, for example “physical”) benefit the least, while domain-dependent relations (e.g., “part-of”) improve the most when encoding domain information.

BibTeX
@inproceedings{bassignana-etal-2024-encode,
    title = "How to Encode Domain Information in Relation Classification",
    author = "Bassignana, Elisa  and
      Gascou, Viggo Unmack  and
      Laustsen, Frida N{\o}hr  and
      Kristensen, Gustav  and
      Petersen, Marie Haahr  and
      van der Goot, Rob  and
      Plank, Barbara",
    editor = "Calzolari, Nicoletta  and
      Kan, Min-Yen  and
      Hoste, Veronique  and
      Lenci, Alessandro  and
      Sakti, Sakriani  and
      Xue, Nianwen",
    booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
    month = may,
    year = "2024",
    address = "Torino, Italia",
    publisher = "ELRA and ICCL",
    url = "https://aclanthology.org/2024.lrec-main.728/",
    pages = "8301--8306"
}
How to Encode Domain Information in Relation Classification · COLING 2024