#natural language process...
Mar 2026
Language Model Maps for Prompt-Response Distributions via Log-Likelihood Vectors
A method that represents language models by log-likelihood vectors over prompt-response pairs and constructs model maps for comparing their conditional distributions is proposed, which supports the analysis and prediction of input-dependent model behavior.
Yusuke Takase, Momose Oyama, H. Shimodaira
· arXiv.org · 0 citations