Imagen de Portada

Sieve: Dynamic Expert-Aware PIM Acceleration for Evolving Mixture-of-Experts Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Kim, Jungwoo, Lacouture, Rubens, Zhang, Genghan, Sohn, Gina, Zhang, Qizheng, Gandhi, Swapnil, Kozyrakis, Christos, Olukotun, Kunle
Formato:	Preprint
Publicado:	2026
Materias:	Hardware Architecture
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Implementing and Optimizing the Scaled Dot-Product Attention on Streaming Dataflow
por: Sohn, Gina, et al.
Publicado: (2024)

Streaming Tensor Programs: A Streaming Abstraction for Dynamic Parallelism
por: Sohn, Gina, et al.
Publicado: (2025)

SSM-RDU: A Reconfigurable Dataflow Unit for Long-Sequence State-Space Models
por: Ko, Sho, et al.
Publicado: (2025)

LowRA: Accurate and Efficient LoRA Fine-Tuning of LLMs under 2 Bits
por: Zhou, Zikai, et al.
Publicado: (2025)

FuseFlow: A Fusion-Centric Compilation Framework for Sparse Deep Learning on Streaming Dataflow
por: Lacouture, Rubens, et al.
Publicado: (2025)

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

AME-PIM: Can Memory be Your Next Tensor Accelerator?
por: Venieri, Emanuele, et al.
Publicado: (2026)

PIM-MMU: A Memory Management Unit for Accelerating Data Transfers in Commercial PIM Systems
por: Lee, Dongjae, et al.
Publicado: (2024)

Revet: A Language and Compiler for Dataflow Threads
por: Rucker, Alexander, et al.
Publicado: (2023)

Inclusive-PIM: Hardware-Software Co-design for Broad Acceleration on Commercial PIM Architectures
por: Alsop, Johnathan, et al.
Publicado: (2023)

LIMINAL: Exploring The Frontiers of LLM Decode Performance
por: Davies, Michael, et al.
Publicado: (2025)

IANUS: Integrated Accelerator based on NPU-PIM Unified Memory System
por: Seo, Minseok, et al.
Publicado: (2024)

Pathfinding Future PIM Architectures by Demystifying a Commercial PIM Technology
por: Hyun, Bongjoon, et al.
Publicado: (2023)

Duplex: A Device for Large Language Models with Mixture of Experts, Grouped Query Attention, and Continuous Batching
por: Yun, Sungmin, et al.
Publicado: (2024)

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

UbiMoE: A Ubiquitous Mixture-of-Experts Vision Transformer Accelerator With Hybrid Computation Pattern on FPGA
por: Dong, Jiale, et al.
Publicado: (2025)

ProactivePIM: Accelerating Weight-Sharing Embedding Layer with PIM for Scalable Recommendation System
por: Kim, Youngsuk, et al.
Publicado: (2024)

LP5X-PIM Sim: A High-Fidelity HW/SW Integrated Simulator for LPDDR5X-PIM
por: Cha, SangHoon, et al.
Publicado: (2026)

PIM-malloc: A Fast and Scalable Dynamic Memory Allocator for Processing-In-Memory (PIM) Architectures
por: Lee, Dongjae, et al.
Publicado: (2025)

A3D-MoE: Acceleration of Large Language Models with Mixture of Experts via 3D Heterogeneous Integration
por: Huang, Wei-Hsing, et al.
Publicado: (2025)

Area-Efficient In-Memory Computing for Mixture-of-Experts via Multiplexing and Caching
por: Gao, Hanyuan, et al.
Publicado: (2026)

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

Membrane: Accelerating Database Analytics with Bank-Level DRAM-PIM Filtering
por: Shekar, Akhil, et al.
Publicado: (2025)

PIM-GPT: A Hybrid Process-in-Memory Accelerator for Autoregressive Transformers
por: Wu, Yuting, et al.
Publicado: (2023)

Context-Aware Mixture-of-Experts Inference on CXL-Enabled GPU-NDP Systems
por: Fan, Zehao, et al.
Publicado: (2025)

SSD Offloading for LLM Mixture-of-Experts Weights Considered Harmful in Energy Efficiency
por: Kyung, Kwanhee, et al.
Publicado: (2025)

Expert Streaming: Accelerating Low-Batch MoE Inference via Multi-chiplet Architecture and Dynamic Expert Trajectory Scheduling
por: Ma, Songchen, et al.
Publicado: (2026)

The BRAM is the Limit: Shattering Myths, Shaping Standards, and Building Scalable PIM Accelerators
por: Kabir, MD Arafat, et al.
Publicado: (2024)

Rethinking LLM Inference Bottlenecks: Insights from Latent Attention and Mixture-of-Experts
por: Yun, Sungmin, et al.
Publicado: (2025)

Annotated PIM Bibliography
por: Kogge, Peter M.
Publicado: (2026)

Scaling Multi-Node Mixture-of-Experts Inference Using Expert Activation Patterns
por: Bambhaniya, Abhimanyu, et al.
Publicado: (2026)

CoQMoE: Co-Designed Quantization and Computation Orchestration for Mixture-of-Experts Vision Transformer on FPGA
por: Dong, Jiale, et al.
Publicado: (2025)

JSPIM: A Skew-Aware PIM Accelerator for High-Performance Databases Join and Select Operations
por: Tajdari, Sabiha, et al.
Publicado: (2025)

LOCALUT: Harnessing Capacity-Computation Tradeoffs for LUT-Based Inference in DRAM-PIM
por: Hong, Junguk, et al.
Publicado: (2026)

Fast-OverlaPIM: A Fast Overlap-driven Mapping Framework for Processing In-Memory Neural Network Acceleration
por: Wang, Xuan, et al.
Publicado: (2024)

HH-PIM: Dynamic Optimization of Power and Performance with Heterogeneous-Hybrid PIM for Edge AI Devices
por: Jeon, Sangmin, et al.
Publicado: (2025)

THERMOS: Thermally-Aware Multi-Objective Scheduling of AI Workloads on Heterogeneous Multi-Chiplet PIM Architectures
por: Kanani, Alish, et al.
Publicado: (2025)

RED: Energy Optimization Framework for eDRAM-based PIM with Reconfigurable Voltage Swing and Retention-aware Scheduling
por: Kim, Jae-Young, et al.
Publicado: (2025)

NASiC: 3D NAND-based CAM-Selected Multibit CIM Architecture for Efficient On-Device Mixture-of-Experts LLM Inference
por: Xu, Weikai, et al.
Publicado: (2026)

SAL-PIM: A Subarray-level Processing-in-Memory Architecture with LUT-based Linear Interpolation for Transformer-based Text Generation
por: Han, Wontak, et al.
Publicado: (2024)