Metraipolis › archive › dc_b4dba9873e4f
JOVE: Joint Execution and Verification for Resource-Aware LLM Task Graphs — arXiv:2610.03296
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.03296
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T22:49:34.444Z
- SHA-256 of the archived text
bbe60384e3c657da70d5dc0ddbcfbac28525be9f930b074c7cd46a29210a8f8d- Stable URL of this page
https://metraipolis.com/documento/dc_b4dba9873e4f
How to cite
JOVE: Joint Execution and Verification for Resource-Aware LLM Task Graphs — arXiv:2610.03296 — arXiv, https://arxiv.org/abs/2610.03296 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b4dba9873e4f (Metraipolis archive dc_b4dba9873e4f, sha256 bbe60384e3c657da70d5dc0ddbcfbac28525be9f930b074c7cd46a29210a8f8d)
BibTeX
@misc{dcb4dba9873e4f,
title = {JOVE: Joint Execution and Verification for Resource-Aware LLM Task Graphs — arXiv:2610.03296},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.03296},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b4dba9873e4f; sha256 bbe60384e3c657da70d5dc0ddbcfbac28525be9f930b074c7cd46a29210a8f8d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03296. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Complex reasoning queries can be decomposed into directed acyclic task graphs and distributed across heterogeneous LLMs, reducing latency through parallelism and enabling smaller models to solve complex tasks. In practice, however, the suitability of an LLM for a given subtask may be a priori unknown, and execution alone does not reveal output correctness. We propose JOVE, an online framework that jointly assigns executor LLMs and selects intermediate outputs for paid verification. Verification runs asynchronously and is used to improve future allocations, so the system must balance spending on execution now against learning for later. We study how to optimize this trade-off under a long-term budget and a per-query latency constraint, with stochastic, initially unknown LLM service quality, invocation costs, and execution times. JOVE makes execution and verification decisions by solving a sequence of per-query mixed-integer linear programs. Online learning updates task-dependent estimates of LLM quality based on verification feedback, while an information-gain bonus incorporates the value of learning into allocation decisions. Under a natural set of assumptions, we establish sublinear quality-learning regret for JOVE. Across four reasoning benchmarks, JOVE achieves competitive accuracy against standard inference baselines while reducing average cost and latency by at least 3.17 times. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.03296 — 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 2026-10-05T22:45:55.685Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).