Someya, Taiga; Svete, Anej; DuSell, Brian; O’Donnell, Timothy J; Giulianelli, Mario; Cotterell, Ryan; (2025) Information Locality as an Inductive Bias for Neural Language Models. In: Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). (pp. pp. 27995-28013). Association for Computational Linguistics

Preview

Text 2025.acl-long.1357.pdf - Published Version Download (1MB) | Preview

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 informationtheoretic measure derived from average lossycontext surprisal—that captures the local uncertainty of a language by quantifying how effectively the m − 1 preceding symbols disambiguate the next symbol. In experiments on both perturbed natural language corpora and languages defined by probabilistic finite-state automata (PFSAs), 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.

Download activity - last month

Export as XMLJSONCSV

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item