Metraipolis › archive › dc_7300bdfeb5e1
Localized TabICLv2: Scaling Tabular In-Context Learning through k-NN — arXiv:2608.16429
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.16429
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T19:41:38.604Z
- SHA-256 of the archived text
ebc39d62aff58a6442f8c445968ed5aea6702b13d9d7675be4b1f06ff33f4d86- Stable URL of this page
https://metraipolis.com/documento/dc_7300bdfeb5e1
How to cite
Localized TabICLv2: Scaling Tabular In-Context Learning through k-NN — arXiv:2608.16429 — arXiv (en), https://arxiv.org/abs/2608.16429 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7300bdfeb5e1 (Metraipolis archive dc_7300bdfeb5e1, sha256 ebc39d62aff58a6442f8c445968ed5aea6702b13d9d7675be4b1f06ff33f4d86)
BibTeX
@misc{dc7300bdfeb5e1,
title = {Localized TabICLv2: Scaling Tabular In-Context Learning through k-NN — arXiv:2608.16429},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.16429},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7300bdfeb5e1; sha256 ebc39d62aff58a6442f8c445968ed5aea6702b13d9d7675be4b1f06ff33f4d86},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.16429. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Foundational models for tabular data have made significant progress in recent years, with TabICLv2 reporting state-of-the-art performance on several tabular classification tasks. However, full-context tabular ICL still suffers from attention cost that grows with the training-context size, which limits its ability to handle large datasets efficiently. Localized TabICLv2 introduces a method that reduces the inference cost of TabICLv2 by retrieving only the k nearest training neighbours for each test point, measured by similarity in the model's Stage 2 row-representation space, rather than using the full training context. This requires no architectural changes, and we show that accuracy retention can be improved through additional Stage 2 and Stage 3 fine-tuning. On TabArena classification tasks, the fine-tuned localized model retains 98.64% of Full TabICLv2 accuracy and it achieves a median 2.18$\times$ speedup in batch inference, and reaches approximately 249$\times$ median speedup in the single-query serving setting. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.16429 — 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 2026-10-18T19:02:44.175Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).