Metraipolis › archive › dc_a68fd895df59

Rethinking Tabular Foundation Models On Data Streams — arXiv:2610.05352

Source
arXiv
Original
https://arxiv.org/abs/2610.05352
License (as archived)
CC0-1.0
Archived
2026-10-06T05:26:51.264Z
SHA-256 of the archived text
8208698653fdfa716924516ff330072816319f11b008b753c847299444cdec91
Stable URL of this page
https://metraipolis.com/documento/dc_a68fd895df59

How to cite

Rethinking Tabular Foundation Models On Data Streams — arXiv:2610.05352 — arXiv, https://arxiv.org/abs/2610.05352 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a68fd895df59 (Metraipolis archive dc_a68fd895df59, sha256 8208698653fdfa716924516ff330072816319f11b008b753c847299444cdec91)
BibTeX
@misc{dca68fd895df59,
  title = {Rethinking Tabular Foundation Models On Data Streams — arXiv:2610.05352},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05352},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a68fd895df59; sha256 8208698653fdfa716924516ff330072816319f11b008b753c847299444cdec91},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05352. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Tabular foundation models (TFMs) outperform established machine learning models on tabular benchmarks through in-context learning. Building on this success, interest is growing in applying them to data streams, where data arrive continuously and evolve over time. On a stream, a TFM adapts by updating its context rather than its parameters, so its accuracy and cost depend on which examples it keeps and how often it rebuilds its context. We therefore present a systematic study of TFMs on data streams, covering memory management, computational cost, and stream-specific challenges such as concept drift and delayed labels. We find that TFMs achieve the highest predictive performance and that simply retaining the most recent examples is as effective as existing memory management techniques. They also recover faster than streaming learners after drift and keep the highest accuracy under label delay. This accuracy, however, comes at a high serving cost, since a nearly unchanged context is re-encoded at every prediction. These results point to architectural efficiency as the way forward for in-context stream learning.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.05352 — 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 2026-10-06T05:24:06.006Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).