COLING 2024main1 citations

Detecting Conceptual Abstraction in LLMs

Michaela Regneri, Alhassan Abdelhalim, Soeren Laue

Abstract

We show a novel approach to detecting noun abstraction within a large language model (LLM). Starting from a psychologically motivated set of noun pairs in taxonomic relationships, we instantiate surface patterns indicating hypernymy and analyze the attention matrices produced by BERT. We compare the results to two sets of counterfactuals and show that we can detect hypernymy in the abstraction mechanism, which cannot solely be related to the distributional similarity of noun pairs. Our findings are a first step towards the explainability of conceptual abstraction in LLMs.

BibTeX
@inproceedings{regneri-etal-2024-detecting,
    title = "Detecting Conceptual Abstraction in {LLM}s",
    author = "Regneri, Michaela  and
      Abdelhalim, Alhassan  and
      Laue, Soeren",
    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.420/",
    pages = "4697--4704"
}