Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Asymptotically Exact Negative Guidance of Diffusion Models via Positive-Unlabeled Learning
Hirohane Takagi, Masashi Sugiyama
Negative guidance of diffusion models generates samples that avoid an undesired condition, specified either by reference samples or by a conditional score alongside an unconditiona…
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…