MVR-cache: Optimizing Semantic Caching via Multi-Vector Retrieval and Learned Prompt Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Noshad, Ali, Zheng, Zishan, Wu, Yinjun
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918520090001408
author Noshad, Ali
Zheng, Zishan
Wu, Yinjun
author_facet Noshad, Ali
Zheng, Zishan
Wu, Yinjun
contents To reduce LLM costs and latency, semantic caching systems must accurately identify when a new prompt matches a cached one. Current methods often rely on simplistic similarity measures, which limit their effectiveness. We introduce MVR-cache, a novel semantic caching approach that significantly improves retrieval accuracy by integrating Multi-Vector Retrieval (MVR). MVR-cache is built upon a learnable segmentation model that intelligently splits prompts, enabling fine-grained similarity comparisons via MaxSim. We derive the model's training objective from a rigorous theoretical analysis. This can ensure that optimizing this objective directly maximizes cache hits under strict correctness constraints. To solve the resulting non-differentiable combinatorial optimization problem, we leverage a reinforcement learning-based training strategy with the theoretically grounded objectives as the reward. Experimental results on established benchmarks across diverse tasks confirm that in comparison to the state-of-the-art, MVR-cache consistently increases the cache hit rates by up to 37% while maintaining the same correctness guarantees. MVR-cache is available at https://github.com/PKU-SDS-lab/MVR-Cache
format Preprint
id arxiv_https___arxiv_org_abs_2605_24914
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MVR-cache: Optimizing Semantic Caching via Multi-Vector Retrieval and Learned Prompt Segmentation
Noshad, Ali
Zheng, Zishan
Wu, Yinjun
Information Retrieval
Databases
Machine Learning
To reduce LLM costs and latency, semantic caching systems must accurately identify when a new prompt matches a cached one. Current methods often rely on simplistic similarity measures, which limit their effectiveness. We introduce MVR-cache, a novel semantic caching approach that significantly improves retrieval accuracy by integrating Multi-Vector Retrieval (MVR). MVR-cache is built upon a learnable segmentation model that intelligently splits prompts, enabling fine-grained similarity comparisons via MaxSim. We derive the model's training objective from a rigorous theoretical analysis. This can ensure that optimizing this objective directly maximizes cache hits under strict correctness constraints. To solve the resulting non-differentiable combinatorial optimization problem, we leverage a reinforcement learning-based training strategy with the theoretically grounded objectives as the reward. Experimental results on established benchmarks across diverse tasks confirm that in comparison to the state-of-the-art, MVR-cache consistently increases the cache hit rates by up to 37% while maintaining the same correctness guarantees. MVR-cache is available at https://github.com/PKU-SDS-lab/MVR-Cache
title MVR-cache: Optimizing Semantic Caching via Multi-Vector Retrieval and Learned Prompt Segmentation
topic Information Retrieval
Databases
Machine Learning
url https://arxiv.org/abs/2605.24914