English
Related papers

Related papers: TileLoom: Automatic Dataflow Planning for Tile-Bas…

200 papers

In recent years, various computing-in-memory (CIM) processors have been presented, showing superior performance over traditional architectures. To unleash the potential of various CIM architectures, such as device precision, crossbar size,…

Hardware Architecture · Computer Science 2024-05-09 Songyun Qu , Shixin Zhao , Bing Li , Yintao He , Xuyi Cai , Lei Zhang , Ying Wang

Computing-in-Memory (CIM) architectures have emerged as a promising solution for accelerating Deep Neural Networks (DNNs) by mitigating data movement bottlenecks. However, realizing the potential of CIM requires specialized dataflow…

Hardware Architecture · Computer Science 2025-10-31 Xiaolin He , Cenlin Duan , Yingjie Qi , Xiao Ma , Jianlei Yang

Multimodal Large Language Models (MLLMs) have achieved remarkable advances by integrating text, image, and audio understanding within a unified architecture. However, existing distributed training frameworks remain fundamentally data-blind:…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-20 Hyeonjun An , Sihyun Kim , Chaerim Lim , Hyunjoon Kim , Rathijit Sen , Sangmin Jung , Hyeonsoo Lee , Dongwook Kim , Takki Yu , Jinkyu Jeong , Youngsok Kim , Kwanghyun Park

In this report, we propose Triton-distributed, an extension of existing Triton compiler, to overcome the programming challenges in distributed AI systems. Triton-distributed is the first compiler that supports native overlapping…

In this paper, we present Hexagon-MLIR,an open-source compilation stack that targets Qualcomm Hexagon Neural Processing Unit (NPU) and provides unified support for lowering Triton kernels and PyTorch models . Built using the MLIR framework,…

Applications with low data reuse and frequent irregular memory accesses, such as graph or sparse linear algebra workloads, fail to scale well due to memory bottlenecks and poor core utilization. While prior work with prefetching,…

Hardware Architecture · Computer Science 2023-05-05 Marcelo Orenes-Vera , Esin Tureci , David Wentzlaff , Margaret Martonosi

Structured LLM workflows, where specialized LLM sub-agents execute according to a predefined graph, have become a powerful abstraction for solving complex tasks. Optimizing such workflows, i.e., selecting configurations for each sub-agent…

Computation and Language · Computer Science 2026-05-14 Junyan Li , Zhang-Wei Hong , Maohao Shen , Yang Zhang , Chuang Gan

The success of DNNs and their high computational requirements pushed for large codesign efforts aiming at DNN acceleration. Since DNNs can be represented as static computational graphs, static memory allocation and tiling are two crucial…

Hardware Architecture · Computer Science 2025-04-08 Victor J. B. Jung , Alessio Burrello , Francesco Conti , Luca Benini

Transformers are central to advances in artificial intelligence (AI), excelling in fields ranging from computer vision to natural language processing. Despite their success, their large parameter count and computational demands challenge…

Hardware Architecture · Computer Science 2025-03-10 Qunyou Liu , Marina Zapater , David Atienza

As the landscape of deep neural networks evolves, heterogeneous dataflow accelerators, in the form of multi-core architectures or chiplet-based designs, promise more flexibility and higher inference performance through scalability. So far,…

Hardware Architecture · Computer Science 2025-10-08 Arne Symons , Linyan Mei , Steven Colleman , Pouya Houshmand , Sebastian Karl , Marian Verhelst

We present Nautilus, a novel tensor compiler that moves toward fully automated math-to-kernel optimization. Nautilus compiles a high-level algebraic specification of tensor operators into efficient tiled GPU kernels. Nautilus's successive…

Programming Languages · Computer Science 2026-04-17 Yifan Zhao , Yuchen Yang , Matei Budiu , Sasa Misailovic

Recently, large language models (LLMs) have achieved huge success in the natural language processing (NLP) field, driving a growing demand to extend their deployment from the cloud to edge devices. However, deploying LLMs on…

Hardware Architecture · Computer Science 2025-05-08 Yanbiao Liang , Huihong Shi , Haikuo Shao , Zhongfeng Wang

Intra-device parallelism addresses resource under-utilization in ML inference and training by overlapping the execution of operators with different resource usage. However, its wide adoption is hindered by a fundamental conflict with the…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-22 Yi Pan , Yile Gu , Jinbin Luo , Yibo Wu , Ziren Wang , Hongtao Zhang , Ziyi Xu , Shengkai Lin , Baris Kasikci , Stephanie Wang

Tensor accelerators now represent a growing share of compute resources in modern CPUs and GPUs. However, they are hard to program, leading developers to use vendor-provided kernel libraries that support tensor accelerators. As a result, the…

Programming Languages · Computer Science 2026-02-12 Yihong Zhang , Derek Gerstmann , Andrew Adams , Maaz Bin Safeer Ahmad

The parameter size of modern large language models (LLMs) can be scaled up via the sparsely-activated Mixture-of-Experts (MoE) technique to avoid excessive increase of the computational costs. To further improve training efficiency,…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-10-08 Yunqi Gao , Bing Hu , Mahdi Boloursaz Mashhadi , A-Long Jin , Yanfeng Zhang , Pei Xiao , Rahim Tafazolli , Merouane Debbah

Spatial dataflow architectures such as reconfigurable dataflow accelerators (RDA) can provide much higher performance and efficiency than CPUs and GPUs. In particular, vectorized reconfigurable dataflow accelerators (vRDA) in recent…

Hardware Architecture · Computer Science 2024-02-01 Alexander Rucker , Shiv Sundram , Coleman Smith , Matthew Vilim , Raghu Prabhakar , Fredrik Kjolstad , Kunle Olukotun

Specialized hardware accelerators are becoming important for more and more applications. Thanks to specialization, they can achieve high performance and energy efficiency but their design is complex and time consuming. This problem is…

Hardware Architecture · Computer Science 2021-04-06 Stephanie Soldavini , Christian Pilato

Bit-serial Processing-In-Memory (PIM) is an attractive paradigm for accelerator architectures, for parallel workloads such as Deep Learning (DL), because of its capability to achieve massive data parallelism at a low area overhead and…

Hardware Architecture · Computer Science 2023-11-21 Aman Arora , Jian Weng , Siyuan Ma , Tony Nowatzki , Lizy K. John

Pipelining between data loading and computation is a critical tensor program optimization for GPUs. In order to unleash the high performance of latest GPUs, we must perform a synergetic optimization of multi-stage pipelining across the…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-05-09 Guyue Huang , Yang Bai , Liu Liu , Yuke Wang , Bei Yu , Yufei Ding , Yuan Xie

The rapidly growing demand for high-quality data in Large Language Models (LLMs) has intensified the need for scalable, reliable, and semantically rich data preparation pipelines. However, current practices remain dominated by ad-hoc…