MIT HAN Lab

AI & ML interests

Efficient algorithm, system, and hardware for machine learning.

Recent Activity

synxlin  authored a paper 14 days ago

Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training

synxlin  authored a paper 14 days ago

TorchSparse: Efficient Point Cloud Inference Engine

synxlin  authored a paper 14 days ago

QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving

View all activity

Papers

VLASH: Real-Time VLAs via Future-State-Aware Asynchronous Inference

View all Papers

Team members 23