Metraipolis › archive › dc_254692c2cb40

Mind the Approximation: Fisher-Weighted SVD Compression for ViTs — arXiv:2609.07155

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07155
License (as archived)
CC0-1.0
Archived
2026-10-16T13:59:07.754Z
SHA-256 of the archived text
afe969d8349c5a25ea71e310ee7d082ce0ff1fef14f7a31a5e505224dbf44cdd
Stable URL of this page
https://metraipolis.com/documento/dc_254692c2cb40

How to cite

Mind the Approximation: Fisher-Weighted SVD Compression for ViTs — arXiv:2609.07155 — arXiv (en), https://arxiv.org/abs/2609.07155 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_254692c2cb40 (Metraipolis archive dc_254692c2cb40, sha256 afe969d8349c5a25ea71e310ee7d082ce0ff1fef14f7a31a5e505224dbf44cdd)
BibTeX
@misc{dc254692c2cb40,
  title = {Mind the Approximation: Fisher-Weighted SVD Compression for ViTs — arXiv:2609.07155},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07155},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_254692c2cb40; sha256 afe969d8349c5a25ea71e310ee7d082ce0ff1fef14f7a31a5e505224dbf44cdd},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07155. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Model compression is key to mitigate deployment challenges of ever growing machine learning models. In this area of research, singular value decomposition (SVD)-based compression offers a compelling trade-off between computational efficiency and model accuracy. Fisher-weighted SVD in particular provides principled, loss-aware compression. However, we find that improving the fidelity of Fisher approximation used in the compression is poorly predictive of post-compression accuracy for Vision Transformers (ViTs). Motivated by this observation, we propose FACTS, a structured Fisher Approximation tailored to Compressing ViTs with Fisher-weighted SVD, which enforces token-local aggregation while preserving within-token activation-gradient dependence. Additionally, we introduce a fast Constrained Rank Search (CoRS), that optimizes layer-wise rank allocation while adhering to a fixed floating point operation (FLOP) constraint. Extensive experiments across ViTs and hybrid architectures demonstrate that FACTS consistently improves accuracy-efficiency trade-offs without requiring finetuning. Notably, it outperforms the strongest SVD baseline by up to +5.8 percentage points (p.p.) Top-1 on Swin-B, with further gains driven by our search method. Code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.LG).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.07155 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T13:39:43.207Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).