Metraipolis › archive › dc_877bd75fd6e8

Tracking States or Tracking Cosets? An Algebraic Account of Learned State Tracking — arXiv:2609.29951

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.29951
License (as archived)
CC0-1.0
Archived
2026-10-08T04:27:53.666Z
SHA-256 of the archived text
e6c15368ed8e54091e6ba706a787615dcfd59954b104b8da3a5f6b7d927dbe34
Stable URL of this page
https://metraipolis.com/documento/dc_877bd75fd6e8

How to cite

Tracking States or Tracking Cosets? An Algebraic Account of Learned State Tracking — arXiv:2609.29951 — arXiv (en), https://arxiv.org/abs/2609.29951 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_877bd75fd6e8 (Metraipolis archive dc_877bd75fd6e8, sha256 e6c15368ed8e54091e6ba706a787615dcfd59954b104b8da3a5f6b7d927dbe34)
BibTeX
@misc{dc877bd75fd6e8,
  title = {Tracking States or Tracking Cosets? An Algebraic Account of Learned State Tracking — arXiv:2609.29951},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29951},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_877bd75fd6e8; sha256 e6c15368ed8e54091e6ba706a787615dcfd59954b104b8da3a5f6b7d927dbe34},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29951. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

State tracking requires composing a sequence of updates, but accuracy alone does not reveal what a model has learned. We study neural networks trained to predict the running product of group elements. We identify quotient solutions in Transformers, where models recover the quotient class while predicting nearly uniformly among its members. The reciprocal of class size predicts partial accuracy without a fitted parameter, extending parity-based accounts to non-parity quotients. Our baseline Transformers' predictions change little under prefix reordering beyond the exact-tracking frontier. We prove that, for finite groups under uniform i.i.d. full-group inputs, optimal order-blind exact accuracy converges to the reciprocal of abelianization class size as prefix length grows, consistent with the observed abelianization plateaus. Sequential updates permit more: any partition into right cosets of a subgroup, normal or not, survives sequential updates. In our census of standard Transformers, every recovered coset partition comes from a normal subgroup, whereas parameter-matched recurrent networks pass through both normal and non-normal right-coset stages during training. On $A_5$, we identify low-dimensional subspaces of the recurrent state that encode non-normal cosets. In the three-dimensional cases, coset mean vectors form approximate dodecahedra, and swapping the state components in these subspaces transfers the donor's coset state through a shared input suffix. Our results connect partial accuracy, learning stages, and internal computation through the subgroup cosets that models learn to track.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.29951 — 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 2026-10-08T03:50:58.192Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).