A Question on the Explainability of Large Language Models and the Word-Level Univariate First-Order Plausibility Assumption

arXiv:2403.10275v1 Announce Type: new
Abstract: The explanations of large language models have recently been shown to be sensitive to the randomness used for their training, creating a need to characterize this sensitivity. In this paper, we propose a characterization that questions the possibility to provide simple and informative explanations for such models. To this end, we give statistical definitions for the explanations' signal, noise and signal-to-noise ratio. We highlight that, in a typical case study where word-level univariate explanations are …

abstract arxiv cs.ai cs.cl explainability language language models large language large language models paper possibility question questions randomness sensitivity training type word