Chia Hong Hsu

Engineer of AI Theories. X for tracking projects I participated in.
Masters @BrownUniversity w. Prof. @randall_balestr & Prof. @jesu9.
Undergrad @hkust.

[Chia Hong Hsu @ch_dinocat35]

This is a work with Co-authors @danielshkao and @randall_balestr. Looking forward to further research on different perspectives of guidance and inversion in generative models.

[Chia Hong Hsu @ch_dinocat35]

Excited to share my first research project—Invertible Guided Consistency Training (iGCT).

iGCT enables fast, deterministic forward & backward mapping along the PFODE path without relying on a diffusion model in training—a first step toward integrating inversion & guidance into Consistency Training.

Website: https://t.co/SuKpq90lhW
Paper: https://t.co/Zxk86AkoLR

[Yang Song @DrYangSong]

Every SDE has an associated probability flow ODE, which yields deterministic processes that sample from the same distribution as the SDE at each timestep. This establishes an equivalence to neural ODEs, allowing sampling via ODE solvers and exact computation of log-likelihoods.