Metraipolis › archive › dc_544fceb18afd

PairCLIP-SWA: Similarity-Guided Balancing and Stochastic Weight Averaging for Generalized Deepfake Image Detection with OpenCLIP — arXiv:2610.37874

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.37874
License (as archived)
CC0-1.0
Archived
2026-10-31T06:11:07.691Z
SHA-256 of the archived text
b66405d74c61a8cb697f09559578087844e04e51f99ca029f4196962df815f27
Stable URL of this page
https://metraipolis.com/documento/dc_544fceb18afd

How to cite

PairCLIP-SWA: Similarity-Guided Balancing and Stochastic Weight Averaging for Generalized Deepfake Image Detection with OpenCLIP — arXiv:2610.37874 — arXiv (en), https://arxiv.org/abs/2610.37874 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_544fceb18afd (Metraipolis archive dc_544fceb18afd, sha256 b66405d74c61a8cb697f09559578087844e04e51f99ca029f4196962df815f27)
BibTeX
@misc{dc544fceb18afd,
  title = {PairCLIP-SWA: Similarity-Guided Balancing and Stochastic Weight Averaging for Generalized Deepfake Image Detection with OpenCLIP — arXiv:2610.37874},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.37874},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_544fceb18afd; sha256 b66405d74c61a8cb697f09559578087844e04e51f99ca029f4196962df815f27},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.37874. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deepfake detectors are almost always deployed as the single checkpoint that scored best on a validation split, yet the reliability of that choice is rarely measured. We measure it with PairCLIP-SWA, a single-backbone detector that adapts an OpenCLIP ViT-L/14 visual tower by progressive block unfreezing on a class-balanced corpus built by CLIP-similarity-guided one-to-one subsampling, and averages the fine-tuning trajectory in weight space. Trained in-domain on the IEEE SP Cup 2025 (DF-Wild-Cup) benchmark under the same protocol as the strongest published three-backbone ensemble, the weight-averaged detector reaches 97.04% accuracy, 98.64% AU-ROC and a 4.07% equal error rate, against 94.46%, 97.60% and 5.00%. Its training log shows why averaging matters: over the averaged epochs the non-averaged checkpoint swings across 5.34 points of validation accuracy while its validation loss more than doubles, whereas the average is never worse and is up to 5.99 points better. On an OpenForensics-derived benchmark the same instability becomes a misranking: the checkpoint that validation accuracy selects (98.43%) scores 88.85% on a disjoint 10,905-image test partition, rejecting 21.1% of genuine photographs, while the terminal average, ranked last on validation, scores 94.76%. A study of twenty-two in-house configurations on the identical DF-Wild-Cup split, including preprocessing ablations, shows that CLIP pre-training rather than ensembling drives accuracy, that secondary face cropping costs about 34 points, and that frequency-domain inputs transfer poorly. We recommend per-class selection criteria measured off the early-stopping split.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.37874 — 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 2026-10-31T06:07:08.601Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).