Continuous Semantic Caching for Low-Cost LLM Serving

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Atalar, Baran, Liu, Xutong, Zuo, Jinhang, Wang, Siwei, Chen, Wei, Joe-Wong, Carlee
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911613947215872
author Atalar, Baran
Liu, Xutong
Zuo, Jinhang
Wang, Siwei
Chen, Wei
Joe-Wong, Carlee
author_facet Atalar, Baran
Liu, Xutong
Zuo, Jinhang
Wang, Siwei
Chen, Wei
Joe-Wong, Carlee
contents As Large Language Models (LLMs) become increasingly popular, caching responses so that they can be reused by users with semantically similar queries has become a vital strategy for reducing inference costs and latency. Existing caching frameworks have proposed to decide which query responses to cache by assuming a finite, known universe of discrete queries and learning their serving costs and arrival probabilities. As LLMs' pool of users and queries expands, however, such an assumption becomes increasingly untenable: real-world LLM queries reside in an infinite, continuous embedding space. In this paper, we establish the first rigorous theoretical framework for semantic LLM response caching in continuous query space under uncertainty. To bridge the gap between discrete optimization and continuous representation spaces, we introduce dynamic $ε$-net discretization coupled with Kernel Ridge Regression. This design enables the system to formally quantify estimation uncertainty and generalize partial feedback on LLM query costs across continuous semantic query neighborhoods. We develop both offline learning and online adaptive algorithms optimized to reduce switching costs incurred by changing the cached responses. We prove that our online algorithm achieves a sublinear regret bound against an optimal continuous oracle, which reduces to existing bounds for discrete query models. Extensive empirical evaluations demonstrate that our framework approximates the continuous optimal cache well while also reducing computational and switching overhead compared to existing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2604_20021
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Continuous Semantic Caching for Low-Cost LLM Serving
Atalar, Baran
Liu, Xutong
Zuo, Jinhang
Wang, Siwei
Chen, Wei
Joe-Wong, Carlee
Machine Learning
Computation and Language
As Large Language Models (LLMs) become increasingly popular, caching responses so that they can be reused by users with semantically similar queries has become a vital strategy for reducing inference costs and latency. Existing caching frameworks have proposed to decide which query responses to cache by assuming a finite, known universe of discrete queries and learning their serving costs and arrival probabilities. As LLMs' pool of users and queries expands, however, such an assumption becomes increasingly untenable: real-world LLM queries reside in an infinite, continuous embedding space. In this paper, we establish the first rigorous theoretical framework for semantic LLM response caching in continuous query space under uncertainty. To bridge the gap between discrete optimization and continuous representation spaces, we introduce dynamic $ε$-net discretization coupled with Kernel Ridge Regression. This design enables the system to formally quantify estimation uncertainty and generalize partial feedback on LLM query costs across continuous semantic query neighborhoods. We develop both offline learning and online adaptive algorithms optimized to reduce switching costs incurred by changing the cached responses. We prove that our online algorithm achieves a sublinear regret bound against an optimal continuous oracle, which reduces to existing bounds for discrete query models. Extensive empirical evaluations demonstrate that our framework approximates the continuous optimal cache well while also reducing computational and switching overhead compared to existing methods.
title Continuous Semantic Caching for Low-Cost LLM Serving
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2604.20021