Metraipolis › archive › dc_95aabdcc6926
Multi-Turn Semantic Caching for Low-Cost LLM Serving — arXiv:2611.01077
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.01077
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:11:45.309Z
- SHA-256 of the archived text
b969863e4a6bb44cc0f2804adc7214c2749b250fc29318016a4c7901c5752c9d- Stable URL of this page
https://metraipolis.com/documento/dc_95aabdcc6926
How to cite
Multi-Turn Semantic Caching for Low-Cost LLM Serving — arXiv:2611.01077 — arXiv (en), https://arxiv.org/abs/2611.01077 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_95aabdcc6926 (Metraipolis archive dc_95aabdcc6926, sha256 b969863e4a6bb44cc0f2804adc7214c2749b250fc29318016a4c7901c5752c9d)
BibTeX
@misc{dc95aabdcc6926,
title = {Multi-Turn Semantic Caching for Low-Cost LLM Serving — arXiv:2611.01077},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.01077},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_95aabdcc6926; sha256 b969863e4a6bb44cc0f2804adc7214c2749b250fc29318016a4c7901c5752c9d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01077. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Semantic caching lowers the cost of serving large language models (LLMs) by reusing stored responses for similar requests. In multi-turn conversations, whether a cached response is acceptable depends on both conversations' histories, which embedding similarity of requests and histories predicts poorly (AUC 0.62--0.74 on five workloads). We therefore learn the \emph{reuse risk}, the probability that a cached response is unacceptable, from judge feedback, and choose a budgeted set of conversation states to cache by weighing the generation cost a reuse avoids against a price for unacceptable reuse. We propose CUCB-SC-MT, which learns a cache from logged conversations with a suboptimality guarantee, and CLCB-LS-MT, which learns online and has $\tilde O(\sqrt N)$ regret over $N$ requests when conversation states recur. On math-tutoring (SocraTeach) and movie-ticketing (Taskmaster-3) conversations, both achieve lower loss than prior semantic caching approaches: ContextCache, multi-turn variants of ElastiCache and vCache, and single-turn semantic caching at every cache budget. With 1,000 cache entries, CLCB-LS-MT avoids 35--41\% of the generation cost through acceptable reuses, against at most 22--25\% for the baselines, and lowers estimated serving latency by 19--35\%.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL); Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2611.01077 — 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 2026-11-03T10:59:52.396Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).