English
Related papers

Related papers: Scalable Processing-Near-Memory for 1M-Token LLM I…

200 papers

Large language models (LLMs) rely on key-value (KV) caches for efficient autoregressive decoding; however, cache size grows linearly with context length and model depth, becoming a major bottleneck in long-context inference. Prior KV cache…

Machine Learning · Computer Science 2025-09-22 Dmitry Akulov , Mohamed Sana , Antonio De Domenico , Tareq Si Salem , Nicola Piovesan , Fadhel Ayed

The rapid adoption of large language models (LLMs) has created significant challenges for efficient inference at scale. Unlike traditional workloads, LLM inference is constrained by both computation and the memory overhead of key-value (KV)…

Machine Learning · Computer Science 2026-05-07 Chengyi Nie , Nian Si , Zijie Zhou

While Large Language Models (LLMs) can theoretically support extensive context windows, their actual deployment is constrained by the linear growth of Key-Value (KV) cache memory. Prevailing compression strategies mitigate this through…

Artificial Intelligence · Computer Science 2026-02-03 Aryan Sood , Tanvi Sharma , Vansh Agrawal

The high computational and memory requirements of large language model (LLM) inference make it feasible only with multiple high-end accelerators. Motivated by the emerging demand for latency-insensitive tasks with batched processing, this…

Large Language Models (LLMs) have made remarkable progress in processing extensive contexts, with the Key-Value (KV) cache playing a vital role in enhancing their performance. However, the growth of the KV cache in response to increasing…

Computation and Language · Computer Science 2024-06-18 Yuhong Li , Yingbing Huang , Bowen Yang , Bharat Venkitesh , Acyr Locatelli , Hanchen Ye , Tianle Cai , Patrick Lewis , Deming Chen

In this work, we design and implement VQ-LLM, an efficient fused Vector Quantization (VQ) kernel generation framework. We first introduce a software abstraction called codebook cache to optimize codebook access efficiency and support the…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-07-01 Zihan Liu , Xinhao Luo , Junxian Guo , Wentao Ni , Yangjie Zhou , Yue Guan , Cong Guo , Weihao Cui , Yu Feng , Minyi Guo , Yuhao Zhu , Minjia Zhang , Jingwen Leng , Chen Jin

Compute Express Link (CXL) serves as a rising industry standard, delivering high-speed cache-coherent links to a variety of devices, including host CPUs, computational accelerators, and memory devices. It is designed to promote system…

Hardware Architecture · Computer Science 2024-11-14 Yuda An , Shushu Yi , Bo Mao , Qiao Li , Mingzhe Zhang , Ke Zhou , Nong Xiao , Guangyu Sun , Xiaolin Wang , Yingwei Luo , Jie Zhang

As the length of input text increases, the key-value (KV) cache in LLMs imposes prohibitive GPU memory costs and limits long-context inference on resource constrained devices. Existing approaches, such as KV quantization and pruning, reduce…

Machine Learning · Computer Science 2025-12-24 Tenghui Li , Guoxu Zhou , Xuyang Zhao , Yuning Qiu , Qibin Zhao

With the advancements in long-context inference capabilities of large language models (LLMs), the KV cache has become one of the foundational components. However, its substantial GPU memory consumption makes KV cache compression a key…

Computation and Language · Computer Science 2025-03-28 Youhui Zuo , Sibo Wei , Chen Zhang , Zhuorui Liu , Wenpeng Lu , Dawei Song

Large language models (LLMs) have shown exceptional performance and vast potential across diverse tasks. However, the deployment of LLMs with high performance in low-resource environments has garnered significant attention in the industry.…

Artificial Intelligence · Computer Science 2024-07-11 Pujiang He , Shan Zhou , Wenhuan Huang , Changqing Li , Duyi Wang , Bin Guo , Chen Meng , Sheng Gui , Weifei Yu , Yi Xie

Compute Express Link (CXL) switch allows memory extension via PCIe physical layer to address increasing demand for larger memory capacities in data centers. However, CXL attached memory introduces 170ns to 400ns memory latency. This becomes…

Hardware Architecture · Computer Science 2025-03-14 Khan Shaikhul Hadi , Naveed Ul Mustafa , Mark Heinrich , Yan Solihin

Language models (LMs) and their extension, vision-language models (VLMs), have achieved remarkable performance across various tasks. However, they still struggle with complex reasoning tasks that require multimodal or multilingual…

Machine Learning · Computer Science 2025-07-09 Wenyi Wu , Zixuan Song , Kun Zhou , Yifei Shao , Zhiting Hu , Biwei Huang

The trend toward specialized processing devices such as TPUs, DPUs, GPUs, and FPGAs has exposed the weaknesses of PCIe in interconnecting these devices and their hosts. Several attempts have been proposed to improve, augment, or downright…

Databases · Computer Science 2024-09-04 Alberto Lerner , Gustavo Alonso

Long-context language models exhibit impressive performance but remain challenging to deploy due to high GPU memory demands during inference. We propose Memory-efficient Offloaded Mini-sequence Inference (MOM), a method that partitions…

Machine Learning · Computer Science 2025-04-18 Junyang Zhang , Tianyi Zhu , Cheng Luo , Anima Anandkumar

Distributed inference of large language models (LLMs) using tensor parallelism can introduce communication overheads of $20$% even over GPUs connected via NVLink, a high-speed GPU interconnect. Several techniques have been proposed to…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-04 Raja Gond , Nipun Kwatra , Ramachandran Ramjee

Transaction processing systems are the crux for modern data-center applications, yet current multi-node systems are slow due to network overheads. This paper advocates for Compute Express Link (CXL) as a network alternative, which enables…

Hardware Architecture · Computer Science 2025-07-24 Zhao Wang , Yiqi Chen , Cong Li , Dimin Niu , Tianchan Guan , Zhaoyang Du , Xingda Wei , Guangyu Sun

In GPU graph analytics, the use of external memory such as the host DRAM and solid-state drives is a cost-effective approach to processing large graphs beyond the capacity of the GPU onboard memory. This paper studies the use of Compute…

Compute Express Link (CXL) is a rapidly emerging coherent interconnect standard that provides opportunities for memory pooling and sharing. Memory sharing is a well-established software feature that improves memory utilization by avoiding…

Emerging Technologies · Computer Science 2024-04-05 Sunita Jain , Nagaradhesh Yeleswarapu , Hasan Al Maruf , Rita Gupta

Large language models (LLMs) are increasingly applied in long-context scenarios such as multi-turn conversations. However, long contexts pose significant challenges for inference efficiency, including high memory overhead from Key-Value…

Computation and Language · Computer Science 2025-12-18 Kuan Lu , Shuhang Lin , Sai Wu , Yichen Yao , Junhan Yang , Huan Li , Wei Chu , Xu Yinghui , Yuan Qi , Gang Chen

Long-context reasoning is a critical capability of large language models (LLMs), enabling applications such as long-document understanding, summarization, and code generation. However, efficient autoregressive inference relies on the…

Computation and Language · Computer Science 2026-04-28 Zahra Dehghanighobadi , Asja Fischer