CVPR 2026
Poster
Weight Space Representation Learning
via Neural Field Adaptation
IVRL · EPFL
Poster session today
Sat, Jun 6, 2026 · 7:45–9:45 PM CEST. Join the Zoom room to chat with me during the CVPR 2026 poster session.
open zoomTL;DR
1
Properly constrained weights exhibit semantic structure and serve as effective data representations.
2
Weight-space geometry correlates strongly with diffusion generation quality.
Video
Poster
Cite this work
@inproceedings{yang2026wsr,
title = {Weight Space Representation Learning via Neural Field Adaptation},
author = {Yang, Zhuoqian and Salzmann, Mathieu and S{\"u}sstrunk, Sabine},
booktitle = {Proceedings of the IEEE/CVF Conference on
Computer Vision and Pattern Recognition (CVPR)},
year = {2026}
}