English
Related papers

Related papers: SPAD: Specialized Prefill and Decode Hardware for …

200 papers

Large language models (LLMs) are increasingly deployed on customer devices. To support them, current devices are adopting SoCs (System on Chip) with NPUs (Neural Processing Unit) installed. Although high performance is expected, LLM…

Hardware Architecture · Computer Science 2025-11-17 Jianyu Wei , Qingtao Li , Shijie Cao , Lingxiao Ma , Zixu Hao , Yanyong Zhang , Xiaoyan Hu , Ting Cao

On-device inference for Large Language Models (LLMs), driven by increasing privacy concerns and advancements of mobile-sized models, has gained significant interest. However, even mobile-sized LLMs (e.g., Gemma-2B) encounter unacceptably…

Artificial Intelligence · Computer Science 2024-12-17 Daliang Xu , Hao Zhang , Liming Yang , Ruiqi Liu , Gang Huang , Mengwei Xu , Xuanzhe Liu

Recent innovations in generative large language models (LLMs) have made their applications and use-cases ubiquitous. This has led to large-scale deployments of these models, using complex, expensive, and power-hungry AI accelerators, most…

Hardware Architecture · Computer Science 2024-05-21 Pratyush Patel , Esha Choukse , Chaojie Zhang , Aashaka Shah , Íñigo Goiri , Saeed Maleki , Ricardo Bianchini

Large language models (LLMs) have been a disruptive innovation in recent years, and they play a crucial role in our daily lives due to their ability to understand and generate human-like text. Their capabilities include natural language…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-10-17 Akrit Mudvari , Yuang Jiang , Leandros Tassiulas

As large language models (LLMs) have shown great success in many tasks, they are used in various applications. While a lot of works have focused on the efficiency of single-LLM application (e.g., offloading, request scheduling, parallelism…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-03-24 Jingzhi Fang , Yanyan Shen , Yue Wang , Lei Chen

Scaling LLM vocabulary is often used to reduce input sequence length and alleviate attention's quadratic cost. Yet, current LLM architectures impose a critical bottleneck to this procedure: the output projection layer scales linearly with…

Computation and Language · Computer Science 2025-07-23 Amit Ben-Artzy , Roy Schwartz

Modern large language models (LLMs) increasingly depends on efficient long-context processing and generation mechanisms, including sparse attention, retrieval-augmented generation (RAG), and compressed contextual memory, to support complex…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-12 Zifan He , Rui Ma , Yizhou Sun , Jason Cong

Recent advancements in speculative decoding have demonstrated considerable speedup across a wide array of large language model (LLM) tasks. Speculative decoding inherently relies on sacrificing extra memory allocations to generate several…

Machine Learning · Computer Science 2025-06-04 Selin Yildirim , Deming Chen

While scaling laws have been continuously validated in large language models (LLMs) with increasing model parameters, the inherent tension between the inference demands of LLMs and the limited resources of edge devices poses a critical…

The efficiency of long-video inference remains a critical bottleneck, mainly due to the dense computation in the prefill stage of Large Multimodal Models (LMMs). Existing methods either compress visual embeddings or apply sparse attention…

Computer Vision and Pattern Recognition · Computer Science 2026-01-30 Yuxiang Huang , Mingye Li , Xu Han , Chaojun Xiao , Weilin Zhao , Ao Sun , Ziqi Yuan , Hao Zhou , Fandong Meng , Zhiyuan Liu

In long-context large language model (LLM) inference, the prefill stage dominates computation due to self-attention over the complete input context. Sparse attention significantly reduces self-attention computation by limiting each token's…

Hardware Architecture · Computer Science 2026-02-25 Rakshith Jayanth , Viktor Prasanna

Modern large language model (LLM) inference has progressively disaggregated to keep pace with growing model sizes and tight TTFT and TPOT service-level objectives: from chunked-prefill aggregation, to prefill-decode (P/D) disaggregation,…

Speculative decoding accelerates large language model (LLM) inference by using a lightweight draft model to propose tokens that are later verified by a stronger target model. While effective in centralized systems, its behavior in…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-18 Jingwei Song , Wanyi Chen , Xinyuan Song , Max , Chris Tong , Gufeng Chen , Tianyi Zhao , Eric Yang , Bill Shi , Lynn Ai

The substantial memory bandwidth and computational demands of large language models (LLMs) present critical challenges for efficient inference. To tackle this, the literature has explored heterogeneous systems that combine neural processing…

Hardware Architecture · Computer Science 2026-05-05 Yuzong Chen , Chao Fang , Xilai Dai , Yuheng Wu , Thierry Tambe , Marian Verhelst , Mohamed S. Abdelfattah

Existing works on large language model (LLM) decomposition mainly focus on improving performance on downstream tasks, but they ignore the poor parallel inference performance when trying to scale up the model size. To mitigate this important…

Computation and Language · Computer Science 2026-04-21 You-Liang Huang , Xinhao Huang , Chengxi Liao , Zeyi Wen

Modern LLM serving now spans multi-stage pipelines including RAG retrieval and KV cache reuse, each with distinct compute, memory, and latency demands. Inference engines expose a large configuration space with no systematic navigation…

Mixed-precision inference techniques reduce the memory and computational demands of Large Language Models (LLMs) by applying hybrid precision formats to model weights, activations, and KV caches. However, existing systems struggle to (i)…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-18 Li Zhang , Youhe Jiang , Guoliang He , Xin Chen , Han Lv , Qian Yao , Ningsheng Ma , Fangcheng Fu , Kai Chen

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

Large Language Models (LLMs) based on autoregressive, decoder-only Transformers generate text one token at a time, where a token represents a discrete unit of text. As each newly produced token is appended to the partial output sequence,…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-05-06 Dimitrios Kafetzis , Ramin Khalili , Iordanis Koutsopoulos

The autoregressive nature of large language models (LLMs) fundamentally limits inference speed, as each forward pass generates only a single token and is often bottlenecked by memory bandwidth. Speculative decoding has emerged as a…

Machine Learning · Computer Science 2025-12-02 Zihao An , Huajun Bai , Ziqiong Liu , Dong Li , Emad Barsoum