Imagen de Portada

T-MAN: Enabling End-to-End Low-Bit LLM Inference on NPUs via Unified Table Lookup

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Wei, Jianyu, Li, Qingtao, Cao, Shijie, Ma, Lingxiao, Hao, Zixu, Zhang, Yanyong, Hu, Xiaoyan, Cao, Ting
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

LUT-DLA: Lookup Table as Efficient Extreme Low-Bit Deep Learning Accelerator
por: Li, Guoyu, et al.
Publicado: (2025)

LUT Tensor Core: A Software-Hardware Co-Design for LUT-Based Low-Bit LLM Inference
por: Mo, Zhiwen, et al.
Publicado: (2024)

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

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

SAIL: SRAM-Accelerated LLM Inference System with Lookup-Table-based GEMV
por: Zhang, Jingyao, et al.
Publicado: (2025)

Hardware Generation and Exploration of Lookup Table-Based Accelerators for 1.58-bit LLM Inference
por: Geens, Robin, et al.
Publicado: (2026)

Huff-LLM: End-to-End Lossless Compression for Efficient LLM Inference
por: Yubeaton, Patrick, et al.
Publicado: (2025)

BitDecoding: Unlocking Tensor Cores for Long-Context LLMs with Low-Bit KV Cache
por: Du, Dayou, et al.
Publicado: (2025)

MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs
por: Wu, Haoran, et al.
Publicado: (2026)

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

FIXME: Towards End-to-End Benchmarking of LLM-Aided Design Verification
por: Wan, Gwok-Waa, et al.
Publicado: (2025)

MVDRAM: Enabling GeMV Execution in Unmodified DRAM for Low-Bit LLM Acceleration
por: Kubo, Tatsuya, et al.
Publicado: (2025)

SLDB: An End-To-End Heterogeneous System-on-Chip Benchmark Suite for LLM-Aided Design
por: Alvanaki, Elisavet Lydia, et al.
Publicado: (2025)

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

elasticAI.explorer: Towards a Unified End-to-End Framework for Hardware-Aware Neural Architecture Search
por: Maman, Natalie, et al.
Publicado: (2026)

When NPUs Are Not Always Faster: A Stage-Level Analysis of Mobile LLM Inference
por: Li, Pu, et al.
Publicado: (2026)

PowerFlow-DNN: Compiler-Directed Fine-Grained Power Orchestration for End-to-End Edge AI Inference
por: Chen, Paul, et al.
Publicado: (2026)

Benchmarking Ultra-Low-Power $μ$NPUs
por: Millar, Josh, et al.
Publicado: (2025)

Pre-gated MoE: An Algorithm-System Co-Design for Fast and Scalable Mixture-of-Expert Inference
por: Hwang, Ranggi, et al.
Publicado: (2023)

SHIELD: A Segmented Hierarchical Memory Architecture for Energy-Efficient LLM Inference on Edge NPUs
por: Zhang, Jintao, et al.
Publicado: (2026)

PIMCOMP: An End-to-End DNN Compiler for Processing-In-Memory Accelerators
por: Sun, Xiaotian, et al.
Publicado: (2024)

Cool-3D: An End-to-End Thermal-Aware Framework for Early-Phase Design Space Exploration of Microfluidic-Cooled 3DICs
por: Wang, Runxi, et al.
Publicado: (2025)

MCBP: A Memory-Compute Efficient LLM Inference Accelerator Leveraging Bit-Slice-enabled Sparsity and Repetitiveness
por: Wang, Huizheng, et al.
Publicado: (2025)

Mamba-X: An End-to-End Vision Mamba Accelerator for Edge Computing Devices
por: Yoon, Dongho, et al.
Publicado: (2025)

Voyager: An End-to-End Framework for Design-Space Exploration and Generation of DNN Accelerators
por: Prabhu, Kartik, et al.
Publicado: (2025)

Table-Lookup MAC: Scalable Processing of Quantised Neural Networks in FPGA Soft Logic
por: Gerlinghoff, Daniel, et al.
Publicado: (2024)

AutoGNN: End-to-End Hardware-Driven Graph Preprocessing for Enhanced GNN Performance
por: Kang, Seungkwan, et al.
Publicado: (2026)

Towards an End-To-End System for Real-Time Gesture Recognition from Surface Vibrations
por: Hettstedt, Florian, et al.
Publicado: (2026)

MCMComm: Hardware-Software Co-Optimization for End-to-End Communication in Multi-Chip-Modules
por: Raj, Ritik, et al.
Publicado: (2025)

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

FASE: FPGA-Assisted Syscall Emulation for Rapid End-to-End Processor Performance Validation
por: Meng, Chengzhen, et al.
Publicado: (2025)

Striking the Balance: GEMM Performance Optimization Across Generations of Ryzen AI NPUs
por: Taka, Endri, et al.
Publicado: (2025)

HSCO-Bench: An Agent-Driven End-to-End Hardware-Software Co-design Benchmark for Systems-on-Chip
por: Tsai, Pei-Huan, et al.
Publicado: (2026)

System-Level Design Space Exploration for High-Level Synthesis under End-to-End Latency Constraints
por: Liao, Yuchao, et al.
Publicado: (2024)

EasyDRAM: An FPGA-based Infrastructure for Fast and Accurate End-to-End Evaluation of Emerging DRAM Techniques
por: Canpolat, Oğuzhan, et al.
Publicado: (2025)

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

ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs
por: Yang, Jinwu, et al.
Publicado: (2026)

T-MAC: CPU Renaissance via Table Lookup for Low-Bit LLM Deployment on Edge
por: Wei, Jianyu, et al.
Publicado: (2024)

ITERA-LLM: Boosting Sub-8-Bit Large Language Model Inference via Iterative Tensor Decomposition
por: Zheng, Keran, et al.
Publicado: (2025)

MARVEL: An End-to-End Framework for Generating Model-Class Aware Custom RISC-V Extensions for Lightweight AI
por: M, Ajay Kumar, et al.
Publicado: (2025)