paper-with-me

Papers

Rethinking Caching for LLM Serving Systems: Beyond Traditional Heuristics

2025-08-26 · Jungwoo Kim, Minsang Kim, Jaeheon Lee, Chanwoo Moon, Heejin Kim, Taeho Hwang, Woosuk Chung, Yeseong Kim, Sungjin Lee arxiv

Serving Large Language Models (LLMs) at scale requires meeting strict Service Level Objectives (SLOs) under severe computational and memory constraints. Nevertheless, traditional caching strategies fall short: exact-matching and prefix caches neglect query semantics, while state-of-the-art semantic caches remain confined to traditional intuitions, offering little conceptual departure. Building on this, we present SISO, a semantic caching system that redefines efficiency for LLM serving. SISO introduces centroid-based caching to maximize coverage with minimal memory, locality-aware replacement to preserve high-value entries, and dynamic thresholding to balance accuracy and latency under varying workloads. Across diverse datasets, SISO delivers up to 1.71$\times$ higher hit ratios and consistently stronger SLO attainment compared to state-of-the-art systems.

📄 PDF Abstract BibTeX arXiv:2508.18736

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Are Optimal Algorithms Still Optimal? Rethinking Sorting in LLM-Based Pairwise Ranking with Batching and Caching

2025-05-30 · Juan Wisznia, Cecilia Bolaños, Juan Tollo, Giovanni Marraffini 외

We introduce a novel framework for analyzing sorting algorithms in pairwise ranking prompting (PRP), re-centering the cost model around LLM inferences rather than traditional pairwise comparisons. While classical metrics…

Beamforming in Wireless Coded-Caching Systems

2023-09-11 · Sneha Madhusudan, Charitha Madapatha, Behrooz Makki, Hao Guo 외

Increased capacity in the access network poses capacity challenges on the transport network due to the aggregated traffic. However, there are spatial and time correlation in the user data demands that could potentially b…

Cost-Efficient Serving of LLM Agents via Test-Time Plan Caching

2025-06-17 · Qizheng Zhang, Michael Wornow, Kunle Olukotun

LLM-based agentic applications have shown increasingly remarkable capabilities in complex workflows but incur substantial costs due to extensive planning and reasoning requirements. Existing LLM caching techniques (like …

Keyword Extraction

Beyond Few-Step Inference: Accelerating Video Diffusion Transformer Model Serving with Inter-Request Caching Reuse

2026-04-06 · Hao Liu, Ye Huang, Chenghuan Huang, Zhenyi Zheng 외 arxiv

Video Diffusion Transformer (DiT) models are a dominant approach for high-quality video generation but suffer from high inference cost due to iterative denoising. Existing caching approaches primarily exploit similarity …

Video Generation

Semantic Caching for Low-Cost LLM Serving: From Offline Learning to Online Adaptation

2025-08-11 · Xutong Liu, Baran Atalar, Xiangxiang Dai, Jinhang Zuo 외 arxiv

Large Language Models (LLMs) are revolutionizing how users interact with information systems, yet their high inference cost poses serious scalability and sustainability challenges. Caching inference responses, allowing t…

Semantic Similarity