Kirato Yoshihara

Gradient Descent towards a big impact.

About

Hi there! I'm Kirato — a builder with research at heart, currently a final-year undergrad focusing on robotic manipulation at The University of Osaka. Research intern at AIST, previously at Preferred Networks.

Research Interests

Foundation Models, Efficient Learning, Representation Learning

Publications
Breaking the Data Barrier: Robust Few-Shot 3D Vessel Segmentation using Foundation Models
Kirato Yoshihara, Yohei Sugawara, Yuta Tokuoka, Lihang Hong
Under Review MICCAI 2026
Essays
April 2026
The Chimera Estimator: Fixing KL Divergence Gradients in High Dimensions
Schulman's k3 estimator has nice forward properties but exploding gradient variance. What if we use k3 for the forward pass and k1 for the backward?
Fun Facts about Me
Mac or Windows?
Mac
Coffee or tea?
Coffee
City or nature?
Nature — a.k.a. Nagano
Favorite editor?
Claude
Best model weights?
model_final_v4_actually_best_20260409_0342_newer_important.pth
Contact