No Train, all Gain: Self-Supervised Gradients Improve Deep Frozen Representations

Walter Simoncini, Andrei Bursuc, Spyridon Gidaris, Yuki M. Asano
2/3/2026

No official code found. Best implementation: WalterSimoncini/fungivision (40 stars, MIT, last commit 2024-10-31). 1 more below.

Semantic Scholar

Code Implementations(2)

WalterSimoncini/fungivisionCommunity66%
MIT

Library implementation of "No Train, all Gain: Self-Supervised Gradients Improve Deep Frozen Representations"

400Python, Jupyter NotebookJul 5, 20241 years agoMIT
computer-visiondeep-learningretrievalself-supervised-learning
WalterSimoncini/no-train-all-gainCommunity62%
MIT

Code for the paper "No Train, all Gain: Self-Supervised Gradients Improve Deep Frozen Representations"

120Shell, PythonJul 11, 20241 years agoMIT

Cite this paper

@article{simoncini2026train,
  title  = {No Train, all Gain: Self-Supervised Gradients Improve Deep Frozen Representations},
  author = {Walter Simoncini and Andrei Bursuc and Spyridon Gidaris and Yuki M. Asano},
  year   = {2026},
  url    = {https://www.semanticscholar.org/paper/b660d2bca87114417097d7512952f96409ecedc6},
  journal = {NEURIPS 2024 2024}
}

Discussion