English
Related papers

Related papers: LLM Serving Optimization with Variable Prefill and…

200 papers

Recurrent large language models that compete with Transformers in language modeling perplexity are emerging at a rapid rate (e.g., Mamba, RWKV). Excitingly, these architectures use a constant amount of memory during inference. However, due…

Computation and Language · Computer Science 2024-07-09 Simran Arora , Aman Timalsina , Aaryan Singhal , Benjamin Spector , Sabri Eyuboglu , Xinyi Zhao , Ashish Rao , Atri Rudra , Christopher Ré

In long-context Large Language Model (LLM) inference, the Time-To-First-Token (TTFT) latency incurred by the prefill stage has become the foremost bottleneck limiting interactive performance and deployment cost. KV Cache reuse offers a…

Hardware Architecture · Computer Science 2026-05-26 Fei li , Song Liu , Yan Liu , Jinhua Cui , Shiqiang Nie , Jinyu Wang , Weiguo Wu

We present Prompt Cache, an approach for accelerating inference for large language models (LLM) by reusing attention states across different LLM prompts. Many input prompts have overlapping text segments, such as system messages, prompt…

Computation and Language · Computer Science 2024-04-26 In Gim , Guojun Chen , Seung-seob Lee , Nikhil Sarda , Anurag Khandelwal , Lin Zhong

In Text-to-SQL tasks, existing LLM-based methods often include extensive database schemas in prompts, leading to long context lengths and increased prefilling latency. While user queries typically focus on recurrent table sets-offering an…

Computation and Language · Computer Science 2026-01-14 Jinbo Su , Yuxuan Hu , Cuiping Li , Hong Chen , Jia Li , Lintao Ma , Jing Zhang

Efficient inference with Large Language Models (LLMs) increasingly relies on Key-Value (KV) caches to store previously computed key and value vectors at each layer. These caches are essential to minimize redundant computation during…

Hardware Architecture · Computer Science 2026-04-08 Oteo Mamo , Olga Kogiou , Hyunjin Yi , Weikuan Yu

Large language models (LLMs) are increasingly utilized in various complex reasoning tasks due to their excellent instruction following capability. However, the model's performance is highly dependent on the open-ended characteristics of the…

Computation and Language · Computer Science 2026-04-28 Zhenzhen Huang , Chaoning Zhang , Fachrina Dewi Puspitasari , Jiaquan Zhang , Yitian Zhou , Shuxu Chen , Yang Yang

Decoding from large language models (LLMs) typically relies on fixed sampling hyperparameters (e.g., temperature, top-p), despite substantial variation in task difficulty and uncertainty across prompts and individual decoding steps. We…

Machine Learning · Computer Science 2026-03-17 Chloe H. Su , Zhe Ye , Samuel Tenka , Aidan Yang , Soonho Kong , Udaya Ghai

In production environments, large language model (LLM) serving is required to meet stringent service-level objectives (SLOs) amid highly variable request patterns. In practice, request lengths follow a long-tail distribution, which gives…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-26 Qipeng Wang

Current large language model (LLM) serving systems, primarily designed for text completion, are neither efficient nor adaptable for increasingly complex LLM applications due to their inflexible design. We propose a new LLM serving system…

Computation and Language · Computer Science 2025-10-30 In Gim , Lin Zhong

Modern LLM serving systems confront inefficient GPU utilization due to the fundamental mismatch between compute-intensive prefill and memory-bound decode phases. While current practices attempt to address this by organizing these phases…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-09-29 Zejia Lin , Hongxin Xu , Guanyi Chen , Zhiguang Chen , Yutong Lu , Xianwei Zhang

Two widely adopted techniques for LLM inference serving systems today are hybrid batching and disaggregated serving. A hybrid batch combines prefill and decode tokens of different requests in the same batch to improve resource utilization…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-21 Amna Masood , Pratishtha Gaur , Nuwan Jayasena

Serving large language models (LLMs) is important for cloud providers, and caching intermediate results (KV\$) after processing each request substantially improves serving throughput and latency. However, there is limited understanding of…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-02-17 Jiahao Wang , Jinbo Han , Xingda Wei , Sijie Shen , Dingyan Zhang , Chenguang Fang , Rong Chen , Wenyuan Yu , Haibo Chen

Disaggregating the prefill and decoding phases represents an effective new paradigm for generative inference of large language models (LLM), which eliminates prefill-decoding interference and optimizes resource allocation. However, it is…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-02-13 Youhe Jiang , Ran Yan , Binhang Yuan

Most vision-language models (VLMs) apply a large language model (LLM) as the decoder, where the response tokens are generated sequentially through autoregression. Therefore, the number of output tokens can be the bottleneck of the…

Computer Vision and Pattern Recognition · Computer Science 2026-04-07 Sixun Dong , Juhua Hu , Steven Li , Wei Wen , Qi Qian

Large Language Models (LLMs) have revolutionized a wide range of domains such as natural language processing, computer vision, and multi-modal tasks due to their ability to comprehend context and perform logical reasoning. However, the…

Artificial Intelligence · Computer Science 2025-07-31 Haoyang Li , Yiming Li , Anxin Tian , Tianhao Tang , Zhanchao Xu , Xuejia Chen , Nicole Hu , Wei Dong , Qing Li , Lei Chen

To schedule LLM inference, the \textit{shortest job first} (SJF) principle is favorable by prioritizing requests with short output lengths to avoid head-of-line (HOL) blocking. Existing methods usually predict a single output length for…

Machine Learning · Computer Science 2026-05-26 Haoyu Zheng , Yongqiang Zhang , Fangcheng Fu , Xiaokai Zhou , Hao Luo , Hongchao Zhu , Yuanyuan Zhu , Hao Wang , Xiao Yan , Jiawei Jiang

Large language models (LLMs) are increasingly deployed as AI agents that operate in short reasoning-action loops, interleaving model computation with external calls. Unlike traditional chat applications, these agentic workloads require…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-03-12 Yuning Zhang , Yan Yan , Nan Yang , Dong Yuan

LLM decoding is bottlenecked for large batches and long contexts by loading the key-value (KV) cache from high-bandwidth memory, which inflates per-token latency, while the sequential nature of decoding limits parallelism. We analyze the…

Machine Learning · Computer Science 2025-05-28 Ted Zadouri , Hubert Strauss , Tri Dao

Applications based on Large Language Models (LLMs) contains a series of tasks to address real-world problems with boosted capability, which have dynamic demand volumes on diverse backends. Existing serving systems treat the resource demands…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-06-19 Yifei Liu , Zuo Gan , Zhenghao Gan , Weiye Wang , Chen Chen , Yizhou Shan , Xusheng Chen , Zhenhua Han , Yifei Zhu , Shixuan Sun , Minyi Guo

We study the problem of routing queries to large language models (LLMs) under cost, GPU resources, and concurrency constraints. Prior per-query routing methods often fail to control batch-level cost, especially under non-uniform or…

Machine Learning · Computer Science 2026-03-31 Jelena Markovic-Voronov , Kayhan Behdin , Yuanda Xu , Zhengze Zhou , Zhipeng Wang , Rahul Mazumder