Joint processing of linguistic properties in brains and language models - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Joint processing of linguistic properties in brains and language models

Abstract

Language models have been shown to be very effective in predicting brain recordings of subjects experiencing complex language stimuli. For a deeper understanding of this alignment, it is important to understand the correspondence between the detailed processing of linguistic information by the human brain versus language models. We investigate this correspondence via a direct approach, in which we eliminate information related to specific linguistic properties in the language model representations and observe how this intervention affects the alignment with fMRI brain recordings obtained while participants listened to a story. We investigate a range of linguistic properties (surface, syntactic, and semantic) and find that the elimination of each one results in a significant decrease in brain alignment. Specifically, we find that syntactic properties (i.e. Top Constituents and Tree Depth) have the largest effect on the trend of brain alignment across model layers. These findings provide clear evidence for the role of specific linguistic information in the alignment between brain and language models, and open new avenues for mapping the joint information processing in both systems. We make the code publicly available.
Fichier principal
Vignette du fichier
2212.08094.pdf (8.67 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04400635 , version 1 (17-01-2024)

Licence

Attribution

Identifiers

  • HAL Id : hal-04400635 , version 1

Cite

Subba Reddy Oota, Manish Gupta, Mariya Toneva. Joint processing of linguistic properties in brains and language models. NeurIPS 2023 - 37th Conference on Neural Information Processing Systems, Neural Information Processing Systems, Dec 2023, New Orleans, United States. ⟨hal-04400635⟩

Collections

CNRS INRIA INRIA2
12 View
5 Download

Share

Gmail Facebook X LinkedIn More