2 papers
cs.LG2026
Alternating Diffusion for Proximal Sampling with Zeroth Order Queries
Hirohane Takagi, Atsushi Nitanda
This work introduces a new approximate proximal sampler that operates solely with zeroth-order information of the potential function. Prior theoretical analyses have revealed that…
cs.AI2025
Interpreting Multi-Attribute Confounding through Numerical Attributes in Large Language Models
Hirohane Takagi, Gouki Minegishi, Shota Kizawa +2
Although behavioral studies have documented numerical reasoning errors in large language models (LLMs), the underlying representational mechanisms remain unclear. We hypothesize th…