@inproceedings{someya-etal-2025-information,
    title = "Information Locality as an Inductive Bias for Neural Language Models",
    author = "Someya, Taiga  and
      Svete, Anej  and
      DuSell, Brian  and
      O{'}Donnell, Timothy J.  and
      Giulianelli, Mario  and
      Cotterell, Ryan",
    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://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.1357/",
    pages = "27995--28013",
    ISBN = "979-8-89176-251-0",
    abstract = "Inductive biases are inherent in every machine learning system, shaping how models generalize from finite data. In the case of neural language models (LMs), debates persist as to whether these biases align with or diverge from human processing constraints. To address this issue, we propose a quantitative framework that allows for controlled investigations into the nature of these biases. Within our framework, we introduce $m$-local entropy{---}an information-theoretic measure derived from average lossy-context surprisal{---}that captures the local uncertainty of a language by quantifying how effectively the preceding symbols disambiguate the next symbol. In experiments on both perturbed natural language corpora and languages defined by probabilistic finite-state automata (PFSA), we show that languages with higher $m$-local entropy are more difficult for Transformer and LSTM LMs to learn. These results suggest that neural LMs, much like humans, are highly sensitive to the local statistical structure of a language."
}