Metraipolis › archive › dc_bb9e6b07e07d

Nous: Learning and Certifying Memory Decisions Before Source Calibration — arXiv:2610.00094

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.00094
License (as archived)
CC0-1.0
Archived
2026-10-06T12:07:51.423Z
SHA-256 of the archived text
1b8b025b159545c580a523a8fce6237ef6efcadd9492cc17423750e10af2dcd9
Stable URL of this page
https://metraipolis.com/documento/dc_bb9e6b07e07d

How to cite

Nous: Learning and Certifying Memory Decisions Before Source Calibration — arXiv:2610.00094 — arXiv (en), https://arxiv.org/abs/2610.00094 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bb9e6b07e07d (Metraipolis archive dc_bb9e6b07e07d, sha256 1b8b025b159545c580a523a8fce6237ef6efcadd9492cc17423750e10af2dcd9)
BibTeX
@misc{dcbb9e6b07e07d,
  title = {Nous: Learning and Certifying Memory Decisions Before Source Calibration — arXiv:2610.00094},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.00094},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bb9e6b07e07d; sha256 1b8b025b159545c580a523a8fce6237ef6efcadd9492cc17423750e10af2dcd9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00094. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agent memory systems update state decisions from reports whose reliability may be unknown. Existing analyses of source estimation do not determine when a policy can be learned or its improvement certified without identifying the reporting channel. We study these three tasks using the same observed records. For a specified hidden Markov family with continuous source uncertainty, decision learning and powered certification have quadratic sample complexity, whereas fixed-precision source estimation has quartic complexity. We characterize a sharp identified interval for policy gain under an unknown shared-background channel and derive finite-sample certificates under bounded history dependence and conditional copying. Independently trained witness regions support general history spaces, and disagreement-conditioned auditing improves power for sparse revisions. For dependent histories, prediction-count-preserving batches cancel the unknown reporting background and admit conditional certificates. A MultiWOZ 2.4 evaluation uses text-processing policies on 1,000 human-written test dialogues with simulated audits. Balanced batches retain 2.26 percentage points of the full candidate's 7.34 percentage-point mean gain and obtain more positive certificates under weak audits. These results establish task-specific information requirements and provide an auditable policy-revision framework for Nous.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.00094 — 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 2026-10-06T10:21:28.021Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).