22.9 A 12nm 18.1TFLOPs/W Sparse Transformer Processor with Entropy-Based Early Exit, Mixed-Precision Predication and Fine-Grained Power Management-Reference-Cited by-同舟云学术

22.9 A 12nm 18.1TFLOPs/W Sparse Transformer Processor with Entropy-Based Early Exit, Mixed-Precision Predication and Fine-Grained Power Management

Published:2023-02-19 Issue: Volume: Page:
ISSN:
Container-title:2023 IEEE International Solid- State Circuits Conference (ISSCC)
language:
Short-container-title:

Author:

Tambe Thierry¹,Zhang Jeff¹,Hooper Coleman¹,Jia Tianyu²,Whatmough Paul N.¹,Zuckerman Joseph³,Santos Maico Cassel Dos³,Loscalzo Erik Jens³,Giri Davide³,Shepard Kenneth³,Carloni Luca³,Rush Alexander⁴,Brooks David¹,Wei Gu-Yeon¹

Affiliation:

1. Harvard University,Cambridge,MA

2. Peking University,Beijing,China

3. Columbia University,New York,NY

4. Cornell University,New York,NY

Funder

DARPA, NSF

Publisher

IEEE

Link

http://xplorestaging.ieee.org/ielx7/10067248/10067251/10067817.pdf?arnumber=10067817

Reference6 articles.

1. A 16-nm SoC for Noise-Robust Speech and NLP Edge AI Inference with Bayesian Sound Source Separation and Attention-Based DNNs;tambe;IEEE JSSC,2022

2. An 8.93 TOPS/W LSTM Recurrent Neural Network Accelerator Featuring Hierarchical Coarse-Grain Sparsity for On-Device Speech Recognition

3. A 28nm 27.5TOPS/W Approximate-Computing-Based Transformer Processor with Asymptotic Sparsity Speculating and Out-of-Order Computing

4. A 17–95.6 TOPS/W Deep Learning Inference Accelerator with Per-Vector Scaled 4-bit Quantization for Transformers in 5nm

5. EdgeBERT: Sentence-Level Energy Optimizations for Latency-Aware Multi-Task NLP Inference;tambe;IEEE Micro,2021

Cited by 7 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. A 28nm Scalable and Flexible Accelerator for Sparse Transformer Models;Proceedings of the 29th ACM/IEEE International Symposium on Low Power Electronics and Design;2024-08-05

2. Survey of Deep Learning Accelerators for Edge and Emerging Computing;Electronics;2024-07-29

3. Towards Cognitive AI Systems: Workload and Characterization of Neuro-Symbolic AI;2024 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS);2024-05-05

4. 20.5 C-Transformer: A 2.6-18.1μJ/Token Homogeneous DNN-Transformer/Spiking-Transformer Processor with Big-Little Network and Implicit Weight Generation for Large Language Models;2024 IEEE International Solid-State Circuits Conference (ISSCC);2024-02-18

5. 14.1 A Software-Assisted Peak Current Regulation Scheme to Improve Power-Limited Inference Performance in a 5nm AI SoC;2024 IEEE International Solid-State Circuits Conference (ISSCC);2024-02-18