unembedding

Tag

Cards List
#unembedding

The Geometry of Ignorance: LLMs Know When to Temper Bayesian Priors

arXiv cs.LG · 2026-09-04 Cached

The paper introduces the 'direction of ignorance' in LLMs' unembedding geometry, which encodes the training corpus's unigram distribution and acts as a Bayesian prior. It demonstrates how this prior is tempered based on context informativeness, providing insights into model behavior.

0 favorites 0 likes
← Back to home

Submit Feedback