Transforming Hidden States into Binary Semantic Features
Computation and Language
2024-10-01 v1
Abstract
Large language models follow a lineage of many NLP applications that were directly inspired by distributional semantics, but do not seem to be closely related to it anymore. In this paper, we propose to employ the distributional theory of meaning once again. Using Independent Component Analysis to overcome some of its challenging aspects, we show that large language models represent semantic features in their hidden states.
Cite
@article{arxiv.2409.19813,
title = {Transforming Hidden States into Binary Semantic Features},
author = {Tomáš Musil and David Mareček},
journal= {arXiv preprint arXiv:2409.19813},
year = {2024}
}