Metraipolis › archive › dc_168eb89ad454

One-Layer Transformer Provably Learns Multiclass One-Nearest Neighbor in Context — arXiv:2609.01311

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01311
License (as archived)
CC0-1.0
Archived
2026-10-17T06:52:22.128Z
SHA-256 of the archived text
f48b5d805481909f2025ab867e2fd0ebd6a80f27fb9b60291c61889bc9cadc18
Stable URL of this page
https://metraipolis.com/documento/dc_168eb89ad454

How to cite

One-Layer Transformer Provably Learns Multiclass One-Nearest Neighbor in Context — arXiv:2609.01311 — arXiv (en), https://arxiv.org/abs/2609.01311 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_168eb89ad454 (Metraipolis archive dc_168eb89ad454, sha256 f48b5d805481909f2025ab867e2fd0ebd6a80f27fb9b60291c61889bc9cadc18)
BibTeX
@misc{dc168eb89ad454,
  title = {One-Layer Transformer Provably Learns Multiclass One-Nearest Neighbor in Context — arXiv:2609.01311},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01311},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_168eb89ad454; sha256 f48b5d805481909f2025ab867e2fd0ebd6a80f27fb9b60291c61889bc9cadc18},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01311. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We extend recent work establishing an equivalence between one-layer transformers and nearest-neighbor classifiers in the binary setting to the multiclass case. By leveraging the simplex encoding, we show that one-layer transformers with an argmax classification head behave identically to a one-nearest-neighbor classifier in the multiclass setting. This closes a gap left by prior work, whose multiclass result relied on a non-standard rounding-based approach rather than the typical argmax head used in practice.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2609.01311 — 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 2026-10-17T06:41:55.853Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).