Imagen de Portada

SnipSnap: A Joint Compression Format and Dataflow Co-Optimization Framework for Efficient Sparse LLM Accelerator Design

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Wu, Junyi, Fang, Chao, Wang, Zhongfeng
Formato:	Preprint
Publicado:	2025
Materias:	Hardware Architecture
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

BETA: Binarized Energy-Efficient Transformer Accelerator at the Edge
por: Ji, Yuhao, et al.
Publicado: (2024)

An Efficient Sparse Hardware Accelerator for Spike-Driven Transformer
por: Li, Zhengke, et al.
Publicado: (2025)

SnapStream: Efficient Long Sequence Decoding on Dataflow Accelerators
por: Li, Jonathan, et al.
Publicado: (2025)

A Dataflow Compiler for Efficient LLM Inference using Custom Microscaling Formats
por: Cheng, Jianyi, et al.
Publicado: (2023)

A Scalable RISC-V Vector Processor Enabling Efficient Multi-Precision DNN Inference
por: Wang, Chuanning, et al.
Publicado: (2024)

SPEED: A Scalable RISC-V Vector Processor Enabling Efficient Multi-Precision DNN Inference
por: Wang, Chuanning, et al.
Publicado: (2024)

A Sparsity-Aware Autonomous Path Planning Accelerator with HW/SW Co-Design and Multi-Level Dataflow Optimization
por: Zhang, Yifan, et al.
Publicado: (2025)

MIREDO: MIP-Driven Resource-Efficient Dataflow Optimization for Computing-in-Memory Accelerator
por: He, Xiaolin, et al.
Publicado: (2025)

LoopTree: Exploring the Fused-layer Dataflow Accelerator Design Space
por: Gilbert, Michael, et al.
Publicado: (2024)

APT-LLM: Exploiting Arbitrary-Precision Tensor Core Computing for LLM Acceleration
por: Ma, Shaobo, et al.
Publicado: (2025)

LP-Spec: Leveraging LPDDR PIM for Efficient LLM Mobile Speculative Inference with Architecture-Dataflow Co-Optimization
por: He, Siyuan, et al.
Publicado: (2025)

VEDA: Efficient LLM Generation Through Voting-based KV Cache Eviction and Dataflow-flexible Accelerator
por: Wang, Zhican, et al.
Publicado: (2025)

FlatAttention: Dataflow and Fabric Collectives Co-Optimization for Efficient Multi-Head Attention on Tile-Based Many-PE Accelerators
por: Zhang, Chi, et al.
Publicado: (2025)

AccLLM: Accelerating Long-Context LLM Inference Via Algorithm-Hardware Co-Design
por: Liang, Yanbiao, et al.
Publicado: (2025)

A High-Throughput Hardware Accelerator for Lempel-Ziv 4 Compression Algorithm
por: Chen, Tao, et al.
Publicado: (2024)

CODO: An Automated Compiler for Comprehensive Dataflow Optimization
por: Zhang, Weichuang, et al.
Publicado: (2026)

Efficient Arbitrary Precision Acceleration for Large Language Models on GPU Tensor Cores
por: Ma, Shaobo, et al.
Publicado: (2024)

LoopLynx: A Scalable Dataflow Architecture for Efficient LLM Inference
por: Zheng, Jianing, et al.
Publicado: (2025)

Multilayer Dataflow: Orchestrate Butterfly Sparsity to Accelerate Attention Computation
por: Wu, Haibin, et al.
Publicado: (2024)

Anda: Unlocking Efficient LLM Inference with a Variable-Length Grouped Activation Data Format
por: Fang, Chao, et al.
Publicado: (2024)

FlatAttention: Dataflow and Fabric Collectives Co-Optimization for Large Attention-Based Model Inference on Tile-Based Accelerators
por: Zhang, Chi, et al.
Publicado: (2026)

Stream: Design Space Exploration of Layer-Fused DNNs on Heterogeneous Dataflow Accelerators
por: Symons, Arne, et al.
Publicado: (2022)

A Composable Dynamic Sparse Dataflow Architecture for Efficient Event-based Vision Processing on FPGA
por: Gao, Yizhao, et al.
Publicado: (2024)

An FPGA-Based Accelerator Enabling Efficient Support for CNNs with Arbitrary Kernel Sizes
por: Wang, Miaoxin, et al.
Publicado: (2024)

Stream-HLS: Towards Automatic Dataflow Acceleration
por: Basalama, Suhail, et al.
Publicado: (2025)

The Immutable Tensor Architecture: A Pure Dataflow Approach for Secure, Energy-Efficient AI Inference
por: Li, Fang
Publicado: (2025)

HASS: Hardware-Aware Sparsity Search for Dataflow DNN Accelerator
por: Yu, Zhewen, et al.
Publicado: (2024)

SIRA: Scaled-Integer Range Analysis for Optimizing FPGA Dataflow Neural Network Accelerators
por: Umuroglu, Yaman, et al.
Publicado: (2025)

AccelCIM: Systematic Dataflow Exploration for SRAM Compute-in-Memory Accelerator
por: Xue, Chenhao, et al.
Publicado: (2026)

DORA: Dataflow-Instruction Orchestration Architecture for DNN Acceleration
por: Chen, Xingzhen, et al.
Publicado: (2026)

Fast Cross-Operator Optimization of Attention Dataflow
por: Chang, Haodong, et al.
Publicado: (2026)

DFModel: Design Space Optimization of Large-Scale Systems Exploiting Dataflow Mappings
por: Ko, Sho, et al.
Publicado: (2024)

A High-Throughput FPGA Accelerator for Lightweight CNNs With Balanced Dataflow
por: Zhao, Zhiyuan, et al.
Publicado: (2024)

FastMamba: A High-Speed and Efficient Mamba Accelerator on FPGA with Accurate Quantization
por: Wang, Aotao, et al.
Publicado: (2025)

StreamTensor: Make Tensors Stream in Dataflow Accelerators for LLMs
por: Ye, Hanchen, et al.
Publicado: (2025)

DataMaestro: A Versatile and Efficient Data Streaming Engine Bringing Decoupled Memory Access To Dataflow Accelerators
por: Yi, Xiaoling, et al.
Publicado: (2025)

LLMulator: Generalizable Cost Modeling for Dataflow Accelerators with Input-Adaptive Control Flow
por: Chang, Kaiyan, et al.
Publicado: (2025)

Dataflow Optimized Reconfigurable Acceleration for FEM-based CFD Simulations
por: Kapetanakis, Anastassis, et al.
Publicado: (2024)

An FPGA-Based Reconfigurable Accelerator for Convolution-Transformer Hybrid EfficientViT
por: Shao, Haikuo, et al.
Publicado: (2024)

Systolic Array Acceleration of Diagonal-Optimized Sparse-Sparse Matrix Multiplication for Efficient Quantum Simulation
por: Su, Yuchao, et al.
Publicado: (2025)