Metraipolis › archive › dc_f07ec44e37e2

Follow the Geometry, Not the Model: Cold Start Semi-Supervised Learning — arXiv:2609.14451

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.14451
License (as archived)
CC0-1.0
Archived
2026-10-15T22:51:10.297Z
SHA-256 of the archived text
f49c753308282d36c00291f97e1208cd72e7dcd667e039250d5b36ce9ba6102a
Stable URL of this page
https://metraipolis.com/documento/dc_f07ec44e37e2

How to cite

Follow the Geometry, Not the Model: Cold Start Semi-Supervised Learning — arXiv:2609.14451 — arXiv (en), https://arxiv.org/abs/2609.14451 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f07ec44e37e2 (Metraipolis archive dc_f07ec44e37e2, sha256 f49c753308282d36c00291f97e1208cd72e7dcd667e039250d5b36ce9ba6102a)
BibTeX
@misc{dcf07ec44e37e2,
  title = {Follow the Geometry, Not the Model: Cold Start Semi-Supervised Learning — arXiv:2609.14451},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.14451},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f07ec44e37e2; sha256 f49c753308282d36c00291f97e1208cd72e7dcd667e039250d5b36ce9ba6102a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.14451. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern semi-supervised learning (SSL) couples pseudo-label generation and classifier training, using the classifier's own confidence to select the pseudo-labels that are then used to update the model. In the cold-start regime, where at most a few labels per class are available, this coupling is ill-posed, since the classifier cannot supervise itself before it has learned. To address this problem, we propose VAST (Veracity-Aware Semi-Supervised Training), which decouples these two stages. Probabilistic beliefs over the unlabeled set are first inferred directly from the geometry of a frozen self-supervised embedding and only then distilled into an inductive classifier. The construction rests on the Veracity Matrix, a kernel-based structure that aggregates label evidence across the data manifold and admits an interpretation as a Dirichlet posterior under a per-observation powered-likelihood model. Additionally, we introduce Veracity Propagation, a self-terminating belief-spreading step that extends coverage beyond the kernel neighborhood of the labeled set. Under a controlled protocol in which all methods receive identical frozen embeddings and labeled sets, VAST outperforms the strongest graph-based SSL baselines at every operating point across three datasets, with statistically significant gains in 7 of 9 comparisons, while producing a deployable inductive classifier rather than requiring transductive graph inference. Compared with end-to-end confidence-gated SSL, we further find that these methods underperform in this setting and, in our experiments, do not consistently exceed labeled-only performance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.14451 — 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 2026-10-15T22:29:03.675Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).