English
Related papers

Related papers: AMMA: A Multi-Chiplet Memory-Centric Architecture …

200 papers

The quadratic computational complexity of MultiHead SelfAttention (MHSA) remains a fundamental bottleneck in scaling Large Language Models (LLMs) for longcontext tasks. While sparse and linearized attention mechanisms attempt to mitigate…

Computation and Language · Computer Science 2025-12-19 Caner Erden

Scaling inference for large language models (LLMs) is increasingly constrained by limited GPU memory, especially due to growing key-value (KV) caches required for long-context generation. While existing approaches offload KV caches to CPU…

Machine Learning · Computer Science 2025-07-08 Weishu Deng , Yujie Yang , Peiran Du , Lingfeng Xiang , Zhen Lin , Chen Zhong , Song Jiang , Hui Lu , Jia Rao

Large Language Models (LLMs) are increasingly deployed on edge devices with Neural Processing Units (NPUs), yet the decode phase remains memory-intensive, limiting performance. Processing-in-Memory (PIM) offers a promising solution, but…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-18 Hai Huang

Data movement in memory-intensive workloads, such as deep learning, incurs energy costs that are over three orders of magnitude higher than the cost of computation. Since these workloads involve frequent data transfers between memory and…

Hardware Architecture · Computer Science 2025-02-05 Bahareh Khabbazan , Marc Riera , Antonio González

RAPID-LLM is a unified performance modeling framework for large language model (LLM) training and inference on GPU clusters. It couples a DeepFlow-based frontend that generates hardware-aware, operator-level Chakra execution traces from an…

The emergence of Phase-Change Memory (PCM) provides opportunities for directly connecting persistent memory to main memory bus. While PCM achieves high read throughput and low standby power, the critical concerns are its poor write…

Hardware Architecture · Computer Science 2020-07-28 Yinjin Fu

The deployment of large language models (LLMs) presents significant challenges due to their enormous memory footprints, low arithmetic intensity, and stringent latency requirements, particularly during the autoregressive decoding stage.…

Hardware Architecture · Computer Science 2025-11-03 Cenlin Duan , Jianlei Yang , Rubing Yang , Yikun Wang , Yiou Wang , Lingkun Long , Yingjie Qi , Xiaolin He , Ao Zhou , Xueyan Wang , Weisheng Zhao

The rapid evolution of Large Language Model (LLM) agents has necessitated robust memory systems to support cohesive long-term interaction and complex reasoning. Benefiting from the strong capabilities of LLMs, recent research focus has…

Artificial Intelligence · Computer Science 2026-04-16 Weiquan Huang , Zixuan Wang , Hehai Lin , Sudong Wang , Bo Xu , Qian Li , Beier Zhu , Linyi Yang , Chengwei Qin

Multi-Head Attention (MHA) is a critical computational kernel in transformer-based AI models. Emerging scalable tile-based accelerator architectures integrate increasing numbers of tightly-packed processing elements (PEs) with tensor units.…

Hardware Architecture · Computer Science 2025-05-27 Chi Zhang , Luca Colagrande , Renzo Andri , Thomas Benz , Gamze Islamoglu , Alessandro Nadalini , Francesco Conti , Yawei Li , Luca Benini

As large language models (LLMs) and visual language models (VLMs) grow in scale and application, attention mechanisms have become a central computational bottleneck due to their high memory and time complexity. While many efficient…

Machine Learning · Computer Science 2025-07-11 Zhengyu Tian , Anantha Padmanaban Krishna Kumar , Hemant Krishnakumar , Reza Rawassizadeh

Processing-in-Memory (PIM) architectures offer promising solutions for efficiently handling AI applications in energy-constrained edge environments. While traditional PIM designs enhance performance and energy efficiency by reducing data…

Hardware Architecture · Computer Science 2025-12-09 Sangmin Jeon , Kangju Lee , Kyeongwon Lee , Woojoo Lee

The quadratic computational complexity of standard attention mechanisms presents a severe scalability bottleneck for LLMs in long-context scenarios. While hybrid attention mechanisms combining Full Attention (FA) and Sparse Attention (SA)…

Machine Learning · Computer Science 2026-04-10 Quantong Qiu , Zhiyi Hong , Yi Yang , Haitian Wang , Kebin Liu , Qingqing Dang , Juntao Li , Min Zhang

As modern LLMs support thousands to millions of tokens, KV caches grow to hundreds of gigabytes, stressing memory capacity and bandwidth. Existing solutions, such as KV cache pruning and offloading, alleviate these but underutilize hardware…

Performance · Computer Science 2026-04-21 Mao Lin , Xi Wang , Guilherme Cox , Dong Li , Hyeran Jeon

Transformer-based large language models (LLMs) excel in modeling complex language patterns but face significant computational costs during inference, especially with long inputs due to the attention mechanism's memory overhead. We observe…

Computation and Language · Computer Science 2024-10-18 Ruiqing Yan , Linghan Zheng , Xingbo Du , Han Zou , Yufeng Guo , Jianfei Yang

Distributed attention is a fundamental problem for scaling context window for Large Language Models (LLMs). The state-of-the-art method, Ring-Attention, suffers from scalability limitations due to its excessive communication traffic. This…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-25 Sirui Chen , Jingji Chen , Siqi Zhu , Ziheng Jiang , Yanghua Peng , Xuehai Qian

Deploying large language models (LLMs) for online inference is often constrained by limited GPU memory, particularly due to the growing KV cache during auto-regressive decoding. Hybrid GPU-CPU execution has emerged as a promising solution…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-16 Jiakun Fan , Yanglin Zhang , Xiangchen Li , Dimitrios S. Nikolopoulos

The computational challenges of Large Language Model (LLM) inference remain a significant barrier to their widespread deployment, especially as prompt lengths continue to increase. Due to the quadratic complexity of the attention…

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

As the foundation of large language models (LLMs), self-attention module faces the challenge of quadratic time and memory complexity with respect to sequence length. FlashAttention accelerates attention computation and reduces its memory…

Machine Learning · Computer Science 2024-09-27 Shimao Chen , Zirui Liu , Zhiying Wu , Ce Zheng , Peizhuang Cong , Zihan Jiang , Yuhan Wu , Lei Su , Tong Yang

Multi-Head Latent Attention (MLA), introduced in DeepSeek-V2, improves the efficiency of large language models by projecting query, key, and value tensors into a compact latent space. This architectural change reduces the KV-cache size and…

Hardware Architecture · Computer Science 2026-04-10 Robin Geens , Marian Verhelst