Metraipolis › archive › dc_5693ec788b2f

ViTAMINS: An Empirical Study of Training Self-Supervised Vision Transformers with Synthetic Hard Negatives — arXiv:2609.01041

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01041
License (as archived)
CC0-1.0
Archived
2026-10-17T06:33:52.861Z
SHA-256 of the archived text
0adf6ac15972d593a95c874d8682f7b801746ed146dfcc96ca892a379c255c00
Stable URL of this page
https://metraipolis.com/documento/dc_5693ec788b2f

How to cite

ViTAMINS: An Empirical Study of Training Self-Supervised Vision Transformers with Synthetic Hard Negatives — arXiv:2609.01041 — arXiv (en), https://arxiv.org/abs/2609.01041 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5693ec788b2f (Metraipolis archive dc_5693ec788b2f, sha256 0adf6ac15972d593a95c874d8682f7b801746ed146dfcc96ca892a379c255c00)
BibTeX
@misc{dc5693ec788b2f,
  title = {ViTAMINS: An Empirical Study of Training Self-Supervised Vision Transformers with Synthetic Hard Negatives — arXiv:2609.01041},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01041},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5693ec788b2f; sha256 0adf6ac15972d593a95c874d8682f7b801746ed146dfcc96ca892a379c255c00},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01041. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We introduce ViTAMINS, a method that integrates synthetic hard negatives into unsupervised vision transformer pretraining to improve representation quality. Our approach is thoroughly benchmarked on ImageNet and transfer learning, image retrieval, copy detection, and image, video segmentation tasks. Notably, our proposed negatives give rise to emergent properties, where learned representations contain explicit information about the semantic content of an image and serve as excellent classifiers (up to +11.3% over baselines). ViTAMINS achieves these benefits through simple modifications to existing contrastive frameworks and outperforms competing methods while being more resource efficient, e.g., our ViT-B surpasses V-JEPA with ViT-L. Our findings motivate reconsidering contrastive learning as a simpler yet powerful alternative to dominant generative and self-distillation approaches.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.01041 — 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 2026-10-17T06:15:12.853Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).