Metraipolis › archive › dc_cc5ff0cd7975

Do LLMs Understand Sequential Structure? A Controlled Study of Inference and Generation — arXiv:2610.04977

Source: arXiv
Original: https://arxiv.org/abs/2610.04977
License (as archived): CC0-1.0
Archived: 2026-10-06T05:24:55.593Z
SHA-256 of the archived text: e978ef5b5f27380cdb928bbb3c66b47b50faf44b68088b51d25823f8800c4d69
Stable URL of this page: https://metraipolis.com/documento/dc_cc5ff0cd7975

How to cite

Do LLMs Understand Sequential Structure? A Controlled Study of Inference and Generation — arXiv:2610.04977 — arXiv, https://arxiv.org/abs/2610.04977 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cc5ff0cd7975 (Metraipolis archive dc_cc5ff0cd7975, sha256 e978ef5b5f27380cdb928bbb3c66b47b50faf44b68088b51d25823f8800c4d69)

BibTeX

@misc{dccc5ff0cd7975,
  title = {Do LLMs Understand Sequential Structure? A Controlled Study of Inference and Generation — arXiv:2610.04977},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04977},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cc5ff0cd7975; sha256 e978ef5b5f27380cdb928bbb3c66b47b50faf44b68088b51d25823f8800c4d69},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04977. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) are increasingly used as interactive agents and simulators, yet it remains unclear whether they can recover latent sequential structure beyond surface action frequencies. This distinction is critical for behavioral simulation, where actions are often shaped by prior context rather than marginal frequencies alone. We study this question using controlled two-player Rock--Paper--Scissors interactions and a one-player stochastic n-gram continuation task. Across these experiments, we test whether LLMs can identify latent strategies, follow simple Markov rules, and sustain higher-order conditional dependencies. Our framework separates distribution matching from conditional rule following. Results show that longer context does not improve identification, correct recognition does not ensure faithful simulation, and higher-order dependencies substantially degrade rule recovery. Apparent behavioral fidelity can therefore mask incorrect generative mechanisms.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.04977 — 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 2026-10-06T05:24:05.539Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).