Metraipolis › archive › dc_8d37c331dbee

ProCAP: Probabilistic Cross-Attentive Prompt Learning for Vision-Language Models — arXiv:2609.30434

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.30434
License (as archived): CC0-1.0
Archived: 2026-10-08T03:12:27.838Z
SHA-256 of the archived text: 87866777f4e88be109ca0346f4a31b144f9ce7d97502e7fd4419c2a595441bd3
Stable URL of this page: https://metraipolis.com/documento/dc_8d37c331dbee

How to cite

ProCAP: Probabilistic Cross-Attentive Prompt Learning for Vision-Language Models — arXiv:2609.30434 — arXiv (en), https://arxiv.org/abs/2609.30434 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8d37c331dbee (Metraipolis archive dc_8d37c331dbee, sha256 87866777f4e88be109ca0346f4a31b144f9ce7d97502e7fd4419c2a595441bd3)

BibTeX

@misc{dc8d37c331dbee,
  title = {ProCAP: Probabilistic Cross-Attentive Prompt Learning for Vision-Language Models — arXiv:2609.30434},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.30434},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8d37c331dbee; sha256 87866777f4e88be109ca0346f4a31b144f9ce7d97502e7fd4419c2a595441bd3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30434. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Pre-trained vision-language models such as CLIP can recognize new categories via prompting, but they often struggle when labeled data are scarce or the test distribution shifts. Prompt learning adapts only a small set of parameters while keeping the backbone frozen, yet many existing multimodal prompt learners couple the visual and textual branches weakly and can be brittle in low-shot regimes. We propose ProCAP, a probabilistic cross-attentive prompt learning framework that improves cross-modal interaction and training stability without updating any CLIP weights: it learns both visual and textual prompt tokens and links them through stacked bidirectional multi-head cross-attention so the two branches refine each other across prompt depth. To reduce overfitting under limited supervision, we parameterize prompt tokens with Gaussian means and variances and regularize them with lightweight KL and L2 penalties, and we further add a compact symmetric InfoNCE head that aligns cross-attended image features with class-level text representations in a shared low-dimensional space. Across few-shot base-to-novel generalization on 11 datasets, cross-dataset transfer, and domain generalization on ImageNet shift benchmarks, ProCAP achieves strong aggregate base-to-novel performance and competitive transfer performance while keeping the CLIP backbone unchanged.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-28 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.30434 — 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 2026-10-08T02:50:19.019Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).