Cover Image

SwiftKV: An Edge-Oriented Attention Algorithm and Multi-Head Accelerator for Fast, Efficient LLM Decoding

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Junming, Zhang, Qinyan, Sun, Huajun, Gao, Feiyang, Hu, Sheng, Nie, Rui, Miao, Xiangshui
Format:	Preprint
Published:	2026
Subjects:	Hardware Architecture
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

TeLLMe: An Energy-Efficient Ternary LLM Accelerator for Prefilling and Decoding on Edge FPGAs
by: Qiao, Ye, et al.
Published: (2025)

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

Kelle: Co-design KV Caching and eDRAM for Efficient LLM Serving in Edge Computing
by: Xia, Tianhua, et al.
Published: (2025)

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

HillInfer: Efficient Long-Context LLM Inference on the Edge with Hierarchical KV Eviction using SmartSSD
by: Sun, He, et al.
Published: (2026)

Titanus: Enabling KV Cache Pruning and Quantization On-the-Fly for LLM Acceleration
by: Chen, Peilin, et al.
Published: (2025)

EdgeLLM: A Highly Efficient CPU-FPGA Heterogeneous Edge Accelerator for Large Language Models
by: Huang, Mingqiang, et al.
Published: (2024)

Salca: A Sparsity-Aware Hardware Accelerator for Efficient Long-Context Attention Decoding
by: Fan, Wang, et al.
Published: (2026)

MemIntelli: A Generic End-to-End Simulation Framework for Memristive Intelligent Computing
by: Zhou, Houji, et al.
Published: (2025)

TeLLMe v2: An Efficient End-to-End Ternary LLM Prefill and Decode Accelerator with Table-Lookup Matmul on Edge FPGAs
by: Qiao, Ye, et al.
Published: (2025)

Designing Efficient LLM Accelerators for Edge Devices
by: Haris, Jude, et al.
Published: (2024)

Oaken: Fast and Efficient LLM Serving with Online-Offline Hybrid KV Cache Quantization
by: Kim, Minsu, et al.
Published: (2025)

EVA: Accelerating LLM Decoding via an Efficient Vector Quantization Architecture
by: Duan, Bowen, et al.
Published: (2026)

PD-Swap: Prefill-Decode Logic Swapping for End-to-End LLM Inference on Edge FPGAs via Dynamic Partial Reconfiguration
by: Zhang, Yifan, et al.
Published: (2025)

Accelerating LLM Inference via Dynamic KV Cache Placement in Heterogeneous Memory System
by: Fang, Yunhua, et al.
Published: (2025)

Swift: A Multi-FPGA Framework for Scaling Up Accelerated Graph Analytics
by: Jaiyeoba, Oluwole, et al.
Published: (2024)

Adaptive KV Cache Reuse for Fast Long-Context LLM Serving
by: li, Fei, et al.
Published: (2026)

VitaLLM: A Versatile and Tiny Accelerator for Mixed-Precision LLM Inference on Edge Devices
by: Lin, Zi-Wei, et al.
Published: (2026)

TENET: An Efficient Sparsity-Aware LUT-Centric Architecture for Ternary LLM Inference On Edge
by: Huang, Zhirui, et al.
Published: (2025)

NX-CGRA: A Programmable Hardware Accelerator for Core Transformer Algorithms on Edge Devices
by: Prasad, Rohit
Published: (2025)

Pushing up to the Limit of Memory Bandwidth and Capacity Utilization for Efficient LLM Decoding on Embedded FPGA
by: Li, Jindong, et al.
Published: (2025)

CD-PIM: A High-Bandwidth and Compute-Efficient LPDDR5-Based PIM for Low-Batch LLM Acceleration on Edge-Device
by: Lin, Ye, et al.
Published: (2026)

Enabling Efficient Hardware Acceleration of Hybrid Vision Transformer (ViT) Networks at the Edge
by: Dumoulin, Joren, et al.
Published: (2025)

FAST-Prefill: FPGA Accelerated Sparse Attention for Long Context LLM Prefill
by: Jayanth, Rakshith, et al.
Published: (2026)

CompAir: Synergizing Complementary PIMs and In-Transit NoC Computation for Efficient LLM Acceleration
by: Li, Hongyi, et al.
Published: (2025)

COBRA: Algorithm-Architecture Co-optimized Binary Transformer Accelerator for Edge Inference
by: Qiao, Ye, et al.
Published: (2025)

VeriCache: Turning Lossy KV Cache into Lossless LLM Inference
by: Yao, Jiayi, et al.
Published: (2026)

CRYPTONITE: Scalable Accelerator Design for Cryptographic Primitives and Algorithms
by: Maheswaran, Karthikeya Sharma, et al.
Published: (2025)

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

From Quarter to All: Accelerating Speculative LLM Decoding via Floating-Point Exponent Remapping and Parameter Sharing
by: Zhao, Yushu, et al.
Published: (2025)

UniCAIM: A Unified CAM/CIM Architecture with Static-Dynamic KV Cache Pruning for Efficient Long-Context LLM Inference
by: Xu, Weikai, et al.
Published: (2025)

Efficient Kernel Mapping and Comprehensive System Evaluation of LLM Acceleration on a CGLA
by: Ando, Takuto, et al.
Published: (2025)

Efficient yet Accurate End-to-End SC Accelerator Design
by: Li, Meng, et al.
Published: (2024)

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

SpecMamba: Accelerating Mamba Inference on FPGA with Speculative Decoding
by: Zhong, Linfeng, et al.
Published: (2025)

Hardware-Centric Analysis of DeepSeek's Multi-Head Latent Attention
by: Geens, Robin, et al.
Published: (2025)

FPGA-Optimized Hardware Accelerator for Fast Fourier Transform and Singular Value Decomposition in AI
by: Ding, Hong, et al.
Published: (2025)

CIMR-V: An End-to-End SRAM-based CIM Accelerator with RISC-V for AI Edge Device
by: and, Yan-Cheng Guo, et al.
Published: (2025)

Hybrid Photonic-digital Accelerator for Attention Mechanism
by: Li, Huize, et al.
Published: (2025)

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