Metraipolis › archive › dc_4aa7537bd70a

Beyond Tier Labels: Role- and Deployment-Dependent Model Substitution in Multi-Call LLM Workflows — arXiv:2608.09155

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.09155
License (as archived)
CC0-1.0
Archived
2026-10-19T06:49:23.367Z
SHA-256 of the archived text
b3ae3fb16df62011ef3e9717579c37566249881b9fda8dc29dc7607331e4fd1a
Stable URL of this page
https://metraipolis.com/documento/dc_4aa7537bd70a

How to cite

Beyond Tier Labels: Role- and Deployment-Dependent Model Substitution in Multi-Call LLM Workflows — arXiv:2608.09155 — arXiv (en), https://arxiv.org/abs/2608.09155 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4aa7537bd70a (Metraipolis archive dc_4aa7537bd70a, sha256 b3ae3fb16df62011ef3e9717579c37566249881b9fda8dc29dc7607331e4fd1a)
BibTeX
@misc{dc4aa7537bd70a,
  title = {Beyond Tier Labels: Role- and Deployment-Dependent Model Substitution in Multi-Call LLM Workflows — arXiv:2608.09155},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.09155},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4aa7537bd70a; sha256 b3ae3fb16df62011ef3e9717579c37566249881b9fda8dc29dc7607331e4fd1a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.09155. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large multi-call LLM systems pose a scientific problem that query-level routing does not capture: the value of a model depends on where it enters a dependent computation and on the deployment that surrounds that call. Existing routers typically decide \emph{where} to spend a stronger model while treating the benefit of the substitution itself as known. We separate these two decisions through a predicate-action factorization and evaluate it in controlled solve-merge-verify workflows spanning 8-64 solve calls and four three-tier model ladders. The resulting evidence reveals a consistent principle beneath apparently conflicting outcomes. On numeric frequency counting, all-strong reduces RMSE from 4.818 to 1.538 in the Mixed Qwen/GPT ladder, whereas the average Qwen-only ordering reverses. Input-matched interventions further show that the same medium-to-strong action has sharply different value across roles and scales. A semantic task-and-contract shift reverses the Mixed ordering again, while allocation ablations distinguish useful sparse placement from under-coverage and indiscriminate escalation. Together, these results establish model substitution as a deployment-conditioned action rather than a property implied by a tier label, and they provide a practical sequence for large-scale workflow routing: calibrate the action, resolve its role-conditioned effect, and then optimize its placement.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-11 in cs.MA).
Categorie arXiv: Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2608.09155 — 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 2026-10-19T06:45:31.685Z dalla pagina https://arxiv.org/catchup/cs.MA/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).