Metraipolis › archive › dc_5c3bcf51dcbf
DynBranch: Speculative Subgraph Reuse for Dynamic Agentic LLM Serving — arXiv:2609.31047
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.31047
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T01:39:40.266Z
- SHA-256 of the archived text
4c0257aa8d7a85c7e7796fba1640f57ed6cf75d9bd00e27993146c3e7b49d346- Stable URL of this page
https://metraipolis.com/documento/dc_5c3bcf51dcbf
How to cite
DynBranch: Speculative Subgraph Reuse for Dynamic Agentic LLM Serving — arXiv:2609.31047 — arXiv (en), https://arxiv.org/abs/2609.31047 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5c3bcf51dcbf (Metraipolis archive dc_5c3bcf51dcbf, sha256 4c0257aa8d7a85c7e7796fba1640f57ed6cf75d9bd00e27993146c3e7b49d346)
BibTeX
@misc{dc5c3bcf51dcbf,
title = {DynBranch: Speculative Subgraph Reuse for Dynamic Agentic LLM Serving — arXiv:2609.31047},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.31047},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5c3bcf51dcbf; sha256 4c0257aa8d7a85c7e7796fba1640f57ed6cf75d9bd00e27993146c3e7b49d346},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31047. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Agentic LLM workflows decide their execution paths at runtime. Downstream computation may be predictable, or may have run before, yet it cannot begin until the model or the user resolves the branch. We call this serialization the branch-resolution barrier. Caching alone does not hide it: the key that identifies a reusable result is not known until then. In this paper, we propose DynBranch, which makes an unresolved branch addressable before it resolves. Its stable coordinate lets candidate subgraphs run during resolution and completed subgraph results be reused across later requests. A two-level controller admits this work when its expected benefit exceeds the load price. DynBranch sits at the model-API boundary and requires no changes to agent harnesses or model execution engines. Across four agentic workloads with Qwen3-32B on 4x H200 GPUs, DynBranch reduces mean latency by up to 32% over each workload's strongest prior system and by 46-66% against a no-reuse floor, while preserving workflow results. The benefit persists across backbone families and on a commodity Qwen3-8B/RTX 4090 deployment. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI). Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.31047 — 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 2026-10-08T01:12:14.526Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).