Metraipolis › archive › dc_4d927f2ea534

Tydra: An Efficient Hybrid Model for Tabular Data — arXiv:2608.21199

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.21199
License (as archived)
CC0-1.0
Archived
2026-10-18T10:31:13.888Z
SHA-256 of the archived text
8e0f8eac7178c361d45d5fa377ec9e512966f2976b510bb03e46dc9bbedf1d06
Stable URL of this page
https://metraipolis.com/documento/dc_4d927f2ea534

How to cite

Tydra: An Efficient Hybrid Model for Tabular Data — arXiv:2608.21199 — arXiv (en), https://arxiv.org/abs/2608.21199 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4d927f2ea534 (Metraipolis archive dc_4d927f2ea534, sha256 8e0f8eac7178c361d45d5fa377ec9e512966f2976b510bb03e46dc9bbedf1d06)
BibTeX
@misc{dc4d927f2ea534,
  title = {Tydra: An Efficient Hybrid Model for Tabular Data — arXiv:2608.21199},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.21199},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4d927f2ea534; sha256 8e0f8eac7178c361d45d5fa377ec9e512966f2976b510bb03e46dc9bbedf1d06},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.21199. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Transformer-based tabular foundation models such as TabPFN achieve strong predictive performance but incur quadratic computational cost with context length. On the other hand, subquadratic SSM-based alternatives such as Hydra trade away accuracy for efficiency. To balance both, we introduce Tydra, a hybrid Transformer-State Space Model (SSM) architecture for tabular in-context learning that interleaves attention and SSM layers. Across 30 OpenML datasets, Tydra reduces inference time by 30% relative to TabPFN while retaining much of its predictive performance. Tydra also outperforms an approximately ten-times-larger Hydra model while providing faster inference. The results indicate that hybrid architectures are a promising direction for tabular foundation models.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-24 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.21199 — 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 2026-10-18T10:27:55.958Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).