Selected Publications

Google Scholar
arXiv 2026

Finite Difference Flow Optimization for RL Post-Training of Text-to-Image Models

Simplified RL post-training for text-to-image flow models

David McAllister, Miika Aittala, Tero Karras, Janne Hellsten, Angjoo Kanazawa, Timo Aila, Samuli Laine

arXiv Blog
ICLR 2026

Flow Matching Policy Gradients

Redirect flow (FM) towards high reward actions (PG)

David McAllister*, Songwei Ge*, Brent Yi*, Chung Min Kim, Ethan Weber, Hongsuk Choi, Haiwen Feng, Angjoo Kanazawa

arXiv Blog
CVPR 2025

Decentralized Diffusion Models

Training diffusion models across isolated compute clusters

David McAllister*, Matthew Tancik, Jiaming Song, Angjoo Kanazawa

arXiv Blog
CoRL 2025

Visual imitation enables contextual humanoid control from monocular videos

CoRL 2025 Best Student Paper Award

Arthur Allshire*, Hongsuk Choi*, Junyi Zhang*, David McAllister*, A. Zhang, C. M. Kim, T. Darrell, P. Abbeel, J. Malik, A. Kanazawa

arXiv Project
Nature Methods 2025
Ring Deconvolution Microscopy result

Ring Deconvolution Microscopy

Fast spatially varying aberration correction for microscopy.

Amit Kohli*, Anastasios Angelopoulos*, David McAllister, E. Whang, S. You, K. Yanny, and L. Waller

arXiv Nature
SIGGRAPH 2023

Nerfstudio: A Modular Framework for Neural Radiance Field Development

Open-source framework for NeRF research and development

Matthew Tancik*, Ethan Weber*, Evonne Ng*, R. Li, B. Yi, J. Kerr, T. Wang, A. Kristoffersen, J. Austin, K. Salahi, A. Ahuja, David McAllister, and A. Kanazawa

arXiv GitHub

Some of my favorite photos

Photo Page
Selected photograph 1
Selected photograph 2
Selected photograph 3
Selected photograph 4
Selected photograph 5
Selected photograph 6