Cover Image

Accelerating LLM Inference with Flexible N:M Sparsity via A Fully Digital Compute-in-Memory Accelerator

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ramachandran, Akshat, Kundu, Souvik, Raha, Arnab, Kundu, Shamik, Mathaikutty, Deepak K., Krishna, Tushar
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence Hardware Architecture
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

FlexNN: A Dataflow-aware Flexible Deep Learning Accelerator for Energy-Efficient Edge Devices
by: Raha, Arnab, et al.
Published: (2024)

SPARQLe: Sub-Precision Activation Representation for Quantized LLM Inference
by: Parvathy, Aradhana Mohan, et al.
Published: (2026)

MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization
by: Ramachandran, Akshat, et al.
Published: (2024)

SafeCiM: Investigating Resilience of Hybrid Floating-Point Compute-in-Memory Deep Learning Accelerators
by: Bhattacharya, Swastik, et al.
Published: (2025)

GraNNite: Enabling High-Performance Execution of Graph Neural Networks on Resource-Constrained Neural Processing Units
by: Das, Arghadip, et al.
Published: (2025)

StruM: Structured Mixed Precision for Efficient Deep Learning Hardware Codesign
by: Wu, Michael, et al.
Published: (2025)

Progressive Gradient Flow for Robust N:M Sparsity Training in Transformers
by: Bambhaniya, Abhimanyu Rajeshkumar, et al.
Published: (2024)

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

AIRCHITECT v2: Learning the Hardware Accelerator Design Space through Unified Representations
by: Seo, Jamin, et al.
Published: (2025)

FusionCIM: Accelerating LLM Inference with Fusion-Driven Computing-in-Memory Architecture
by: Xuan, Zihao, et al.
Published: (2026)

SOFA: A Compute-Memory Optimized Sparsity Accelerator via Cross-Stage Coordinated Tiling
by: Wang, Huizheng, et al.
Published: (2024)

OneDSE: A Unified Microprocessor Metric Prediction and Design Space Exploration Framework
by: Raj, Ritik, et al.
Published: (2025)

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

TYTAN: Taylor-series based Non-Linear Activation Engine for Deep Learning Accelerators
by: Pramanik, Soham, et al.
Published: (2025)

CiMNet: Towards Joint Optimization for DNN Architecture and Configuration for Compute-In-Memory Hardware
by: Kundu, Souvik, et al.
Published: (2024)

HCiM: ADC-Less Hybrid Analog-Digital Compute in Memory Accelerator for Deep Learning Workloads
by: Negi, Shubham, et al.
Published: (2024)

Hardware-Software Co-Design for Accelerating Transformer Inference Leveraging Compute-in-Memory
by: Kim, Dong Eun, et al.
Published: (2025)

PRIMAL: Processing-In-Memory Based Low-Rank Adaptation for LLM Inference Accelerator
by: Chong, Yue Jiet, et al.
Published: (2026)

MINISA: Minimal Instruction Set Architecture for Next-gen Reconfigurable Inference Accelerator
by: Tong, Jianming, et al.
Published: (2026)

FEATHER: A Reconfigurable Accelerator with Data Reordering Support for Low-Cost On-Chip Dataflow Switching
by: Tong, Jianming, et al.
Published: (2024)

Modeling Analog-Digital-Converter Energy and Area for Compute-In-Memory Accelerator Design
by: Andrulis, Tanner, et al.
Published: (2024)

A Switch-Centric In-Network Architecture for Accelerating LLM Inference in Shared-Memory Network
by: Jiang, Aojie, et al.
Published: (2026)

PICNIC: Silicon Photonic Interconnected Chiplets with Computational Network and In-memory Computing for LLM Inference Acceleration
by: Chong, Yue Jiet, et al.
Published: (2025)

Algorithm-Hardware Co-Design of Distribution-Aware Logarithmic-Posit Encodings for Efficient DNN Inference
by: Ramachandran, Akshat, et al.
Published: (2024)

In-Pipeline Integration of Digital In-Memory-Computing into RISC-V Vector Architecture to Accelerate Deep Learning
by: Spagnolo, Tommaso, et al.
Published: (2026)

RCW-CIM: A Digital CIM-based LLM Accelerator with Read-Compute/Write
by: Guo, Yan-Cheng, et al.
Published: (2026)

Cross-Layer Design of Vector-Symbolic Computing: Bridging Cognition and Brain-Inspired Hardware Acceleration
by: Du, Shuting, et al.
Published: (2025)

ADS-IMC: Accelerating Data Sorting with In-Memory Computation
by: Dhakad, Narendra Singh, et al.
Published: (2026)

FPGA Co-Design for Efficient N:M Sparse and Quantized Model Inference
by: Hsieh, Fen-Yu, et al.
Published: (2025)

Prosperity: Accelerating Spiking Neural Networks via Product Sparsity
by: Wei, Chiyue, et al.
Published: (2025)

HPIM: Heterogeneous Processing-In-Memory-based Accelerator for Large Language Models Inference
by: Duan, Cenlin, et al.
Published: (2025)

Memory-Guided Unified Hardware Accelerator for Mixed-Precision Scientific Computing
by: Wang, Chuanzhen, et al.
Published: (2026)

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

Demystifying AI Platform Design for Distributed Inference of Next-Generation LLM models
by: Bambhaniya, Abhimanyu, et al.
Published: (2024)

NeuPIMs: NPU-PIM Heterogeneous Acceleration for Batched LLM Inferencing
by: Heo, Guseul, et al.
Published: (2024)

VIKIN: A Reconfigurable Accelerator for KANs and MLPs with Two-Stage Sparsity Support
by: Ou, Wenhui, et al.
Published: (2026)

Memory Is All You Need: An Overview of Compute-in-Memory Architectures for Accelerating Large Language Model Inference
by: Wolters, Christopher, et al.
Published: (2024)

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

CIMPool: Scalable Neural Network Acceleration for Compute-In-Memory using Weight Pools
by: Li, Shurui, et al.
Published: (2025)

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