ACL 2025long0 citations

Mapping 1,000+ Language Models via the Log-Likelihood Vector

Momose Oyama, Hiroaki Yamagiwa, Yusuke Takase, Hidetoshi Shimodaira

Abstract

To compare autoregressive language models at scale, we propose using log-likelihood vectors computed on a predefined text set as model features. This approach has a solid theoretical basis: when treated as model coordinates, their squared Euclidean distance approximates the Kullback-Leibler divergence of text-generation probabilities. Our method is highly scalable, with computational cost growing linearly in both the number of models and text samples, and is easy to implement as the required features are derived from cross-entropy loss. Applying this method to over 1,000 language models, we constructed a “model map,” providing a new perspective on large-scale model analysis.

BibTeX
@inproceedings{oyama-etal-2025-mapping,
    title = "Mapping 1,000+ Language Models via the Log-Likelihood Vector",
    author = "Oyama, Momose  and
      Yamagiwa, Hiroaki  and
      Takase, Yusuke  and
      Shimodaira, Hidetoshi",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.acl-long.1584/",
    doi = "10.18653/v1/2025.acl-long.1584",
    pages = "32983--33038",
    ISBN = "979-8-89176-251-0"
}