Explore the groundbreaking research presented in this 19-minute conference talk video from MLSys 2024, featuring the Best Paper "AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration." Delve into the innovative approach developed by researchers from MIT HAN Lab for compressing and accelerating Large Language Models (LLMs). Learn about the Activation-aware Weight Quantization (AWQ) technique and its potential impact on improving the efficiency of LLMs. Gain insights into the methodology, results, and implications of this cutting-edge work in machine learning systems. Access additional resources, including the project website, full paper, and code repository, to further understand and potentially implement the AWQ technique in your own projects.
AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
MIT HAN Lab via YouTube
Overview
Syllabus
MLSys'24 Best Paper - AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
Taught by
MIT HAN Lab
Related Courses
-
AWQ for LLM Quantization - Efficient Inference Framework for Large Language Models
-
ELL: The Microsoft Embedded Learning Library - Principles and Applications
-
Faster and Cheaper LLMs with Weight and Key-value Cache Quantization
-
Compressing Large Language Models (LLMs) with Python Code - 3 Techniques
-
SWIS: Shared Weight Bit Sparsity for Efficient Neural Network Acceleration
-
Unlock Faster and More Efficient LLMs with SparseGPT - Neural Magic
