Metraipolis › archive › dc_80564d1e201d

Measurement-Driven Sub-Network Selection for On-Premise Retrieval-Augmented Factory Agents — arXiv:2609.02760

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.02760
License (as archived)
CC0-1.0
Archived
2026-10-17T03:26:38.090Z
SHA-256 of the archived text
e99a01e0a663a4810838fa768616b992fa3a7f07397719f322935380d9906413
Stable URL of this page
https://metraipolis.com/documento/dc_80564d1e201d

How to cite

Measurement-Driven Sub-Network Selection for On-Premise Retrieval-Augmented Factory Agents — arXiv:2609.02760 — arXiv (en), https://arxiv.org/abs/2609.02760 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_80564d1e201d (Metraipolis archive dc_80564d1e201d, sha256 e99a01e0a663a4810838fa768616b992fa3a7f07397719f322935380d9906413)
BibTeX
@misc{dc80564d1e201d,
  title = {Measurement-Driven Sub-Network Selection for On-Premise Retrieval-Augmented Factory Agents — arXiv:2609.02760},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.02760},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_80564d1e201d; sha256 e99a01e0a663a4810838fa768616b992fa3a7f07397719f322935380d9906413},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.02760. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

On-premise assistants can give factory workers conversational access to machine documentation, but models capable of the task rarely fit shop-floor hardware. We show that after structural compression and retrieval-grounded adaptation, model size is no longer a reliable predictor of adapted answer quality: general capability falls almost linearly with parameter count, while judged retrieval-augmented answer quality does not. We therefore treat deployment as a post-adaptation selection problem, committing one sub-network per device on judged answer quality and measured on-device throughput under a configurable general-capability floor and memory budget; rules that optimize size, speed, or quality alone each give up capability or throughput. A weight-shared supernetwork trained with sandwich-style in-place distillation keeps this selection inexpensive. In a manufacturing-manual case study, extraction costs 13.7 percent of the unpruned model's judged quality and retrieval-grounded distillation returns it to within 4.6 percent, recovering two thirds of the loss, and the same assistant runs across three heterogeneous edge tiers at 1.3 to 5 watts standby.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-03 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.02760 — 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 2026-10-17T03:13:12.451Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).