Metraipolis › archive › dc_e4789a4eba67

Catching the Imposter: Self-Supervised Learning of Physical Coherence with Cross-Entity Feature Permutations — arXiv:2608.14372

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.14372
License (as archived): CC0-1.0
Archived: 2026-10-19T02:30:39.668Z
SHA-256 of the archived text: 0f02a819110f89267731f3f73086f643f215b889c4c60ae5aaf3337cc60c5af6
Stable URL of this page: https://metraipolis.com/documento/dc_e4789a4eba67

How to cite

Catching the Imposter: Self-Supervised Learning of Physical Coherence with Cross-Entity Feature Permutations — arXiv:2608.14372 — arXiv (en), https://arxiv.org/abs/2608.14372 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e4789a4eba67 (Metraipolis archive dc_e4789a4eba67, sha256 0f02a819110f89267731f3f73086f643f215b889c4c60ae5aaf3337cc60c5af6)

BibTeX

@misc{dce4789a4eba67,
  title = {Catching the Imposter: Self-Supervised Learning of Physical Coherence with Cross-Entity Feature Permutations — arXiv:2608.14372},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14372},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e4789a4eba67; sha256 0f02a819110f89267731f3f73086f643f215b889c4c60ae5aaf3337cc60c5af6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14372. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Scientific data often describe entities whose features are jointly governed by the laws of physics, yet existing self-supervised learning (SSL) objectives largely ignore this physical coherence. We introduce imposter, a discriminative pretext task that replaces subsets of an entity's features with real observations donated by another entity and trains the encoder to identify the swapped features. Because every donated value is individually plausible, the task can only be solved by learning cross-feature physical dependencies. We evaluate the proposed objectives on global ERA5-Land reanalysis data using 21 environmental variables and assess the learned representations on seven downstream tasks spanning climate classification, carbon flux estimation, and streamflow prediction. Our study includes, to our knowledge, the first systematic comparison of self-supervised objectives for land-surface modeling under a shared architecture and pre-training budget. We find that the most effective pretext task depends on the downstream task family rather than any single objective's superiority, and that imposter provides complementary information when combined with existing SSL objectives. These results suggest that physical coherence is a valuable new source of self-supervision for scientific foundation models.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.14372 — 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 2026-10-19T02:26:17.805Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).