Metraipolis › archive › dc_c7356fbfa4b4

Pretraining Reusable Inference Across Views with Synthetic Task Priors — arXiv:2608.19115

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.19115
License (as archived)
CC0-1.0
Archived
2026-10-18T13:53:34.326Z
SHA-256 of the archived text
d713b9473217adfb5ea20f3ac3f3f26d5a22ccf3fe6157b6126ea1734825be71
Stable URL of this page
https://metraipolis.com/documento/dc_c7356fbfa4b4

How to cite

Pretraining Reusable Inference Across Views with Synthetic Task Priors — arXiv:2608.19115 — arXiv (en), https://arxiv.org/abs/2608.19115 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c7356fbfa4b4 (Metraipolis archive dc_c7356fbfa4b4, sha256 d713b9473217adfb5ea20f3ac3f3f26d5a22ccf3fe6157b6126ea1734825be71)
BibTeX
@misc{dcc7356fbfa4b4,
  title = {Pretraining Reusable Inference Across Views with Synthetic Task Priors — arXiv:2608.19115},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.19115},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c7356fbfa4b4; sha256 d713b9473217adfb5ea20f3ac3f3f26d5a22ccf3fe6157b6126ea1734825be71},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19115. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern pretrained encoders make representations from heterogeneous views increasingly reusable, but the procedure that determines view utility and combines evidence is still relearned for each downstream task. Consequently, knowledge about view relevance, complementarity, reliability, and missingness is repeatedly discarded rather than transferred across tasks. We therefore reformulate multi-view learning as learning a reusable, task-conditioned inference procedure rather than a fixed fusion function. Based on this perspective, we propose SIMPLE, a prior-fitted multi-view in-context learner that predicts query labels by conditioning on a small labeled support set. Since existing real-world datasets cover only a limited range of view configurations and task structures, we construct a controllable synthetic task prior in embedding space. It generates diverse support-query episodes with varying class structures, shared and view-specific factors, representation geometries, cross-view dependencies, reliability levels, missingness patterns, and distribution shifts. A hierarchical inference architecture then performs reasoning within views, across views, and across support and query samples. Experiments on multi-view and multi-omics benchmarks demonstrate that the frozen variant of SIMPLE achieves competitive performance without updating the inference backbone, while lightweight adapter calibration attains leading performance on most evaluated datasets. Together, the results under frozen, one-shot, and missing-view settings support the central hypothesis that multi-view reasoning itself can be pretrained and reused, while lightweight adapter calibration provides task-specific alignment when needed.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-20 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Multimedia (cs.MM).
Pagina del lavoro: https://arxiv.org/abs/2608.19115 — 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 2026-10-18T13:50:58.961Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).