Metraipolis › archive › dc_c2287e57d577
Support-Compiled Feature Folding: More Evidence at Lower Memory Across Tabular Foundation Models — arXiv:2609.28208
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.28208
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T10:57:38.232Z
- SHA-256 of the archived text
044fe3e4c36275caabde448750c3195ed8a831accd77a67a8487f43d39ba76ef- Stable URL of this page
https://metraipolis.com/documento/dc_c2287e57d577
How to cite
Support-Compiled Feature Folding: More Evidence at Lower Memory Across Tabular Foundation Models — arXiv:2609.28208 — arXiv (en), https://arxiv.org/abs/2609.28208 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c2287e57d577 (Metraipolis archive dc_c2287e57d577, sha256 044fe3e4c36275caabde448750c3195ed8a831accd77a67a8487f43d39ba76ef)
BibTeX
@misc{dcc2287e57d577,
title = {Support-Compiled Feature Folding: More Evidence at Lower Memory Across Tabular Foundation Models — arXiv:2609.28208},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.28208},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c2287e57d577; sha256 044fe3e4c36275caabde448750c3195ed8a831accd77a67a8487f43d39ba76ef},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.28208. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Wide tables offer tabular foundation models more evidence, but accessing it can exhaust their memory: full-width pairwise mixing grows quadratically with the number of columns, while feature selection makes inputs affordable by discarding evidence. We ask whether using more features requires interacting over all of them at once. We introduce Support-Compiled Feature Folding (SCFF), a training-free inference framework that encodes wide tables through bounded calls to a frozen backbone. SCFF organizes support-ranked features into a strong Core and a candidate Tail, folds them into narrow feature groups, and support-checks the Tail's added evidence before a single contextual prediction. This converts quadratic feature-interaction work into linear-in-width work with a bounded local working set, without ensembling predictions or training new parameters. On the exhaustive 18-dataset wide-table slice of fixed AMLB-29, TabZilla, and TabArena snapshots, SCFF improves dataset-macro accuracy and NLL on all six evaluated backbones. All four matched-width comparisons retain favorable 95% dataset-bootstrap intervals on locked folds, with relative error reductions up to 26.1%. Median paired GPU-memory savings are 2.09-2.36x, and the ratio of separately observed maximum peaks reaches 34.3x. Under a measured peak-memory ceiling, SCFF uses the saved budget to preserve more support-selected evidence, improving accuracy by 4.06 and 3.72 points over the widest feasible single leaf on predeclared wide-Core strata of TabICLv2 and TabPFN-3. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.28208 — 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 2026-10-08T10:06:16.777Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).