Metraipolis › archive › dc_8ab72e60045d
GLF-Q: Global-Local Feature-based Quantization for Vision Transformers — arXiv:2609.34564
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34564
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T23:52:16.113Z
- SHA-256 of the archived text
d114e2a0b0a4bd57a5506c7ea277c652016806ad8c2c62496a61432772521ae0- Stable URL of this page
https://metraipolis.com/documento/dc_8ab72e60045d
How to cite
GLF-Q: Global-Local Feature-based Quantization for Vision Transformers — arXiv:2609.34564 — arXiv (en), https://arxiv.org/abs/2609.34564 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8ab72e60045d (Metraipolis archive dc_8ab72e60045d, sha256 d114e2a0b0a4bd57a5506c7ea277c652016806ad8c2c62496a61432772521ae0)
BibTeX
@misc{dc8ab72e60045d,
title = {GLF-Q: Global-Local Feature-based Quantization for Vision Transformers — arXiv:2609.34564},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34564},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8ab72e60045d; sha256 d114e2a0b0a4bd57a5506c7ea277c652016806ad8c2c62496a61432772521ae0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34564. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Post-training quantization (PTQ) efficiently compresses Vision Transformers (ViTs) without retraining, yet suffers severe accuracy degradation at low bit-widths. Existing optimization-based PTQ methods guide block reconstruction via either soft logits or second-order Hessian proxies. Logit supervision is prone to overfitting on limited calibration data, while Hessian approximations incur structural truncation errors. To address these limitations, we propose \textbf{GLF-Q}, a novel PTQ framework guided by Global-Local Feature alignment. GLF-Q propagates quantized block outputs through downstream full-precision layers to align penultimate-layer representations under local output regularization, providing downstream feature supervision without explicitly approximating the Hessian or using a Taylor expansion. Furthermore, offline Hadamard transformations are introduced with zero runtime overhead to disperse activation outliers across channels, effectively contracting dynamic ranges and reducing quantization errors. Meanwhile, optimizing this loss via a Straight-Through Estimator (STE) achieves rapid convergence, bypassing continuous relaxation rounding formulations such as AdaRound. Extensive experiments across representative ViT architectures demonstrate that GLF-Q with standard uniform quantizers substantially outperforms state-of-the-art methods under 3-bit quantization on image classification. In addition, GLF-Q exhibits strong out-of-domain calibration robustness and achieves speedups under 8-bit GPU deployment.
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.34564 — 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 2026-10-07T23:29:27.395Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).