Song Han
Source-listed: Associate Professor with tenure, Department of Electrical Engineering and Computer Science (EECS); Principal Investigator, MIT HAN Lab · Massachusetts Institute of Technology (MIT)
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman CodingSaved credit: First author · 2016Explore people connected to this work →
- EIE: Efficient Inference Engine on Compressed Deep Neural NetworkSaved credit: First author · 2016Explore people connected to this work →
- Once-for-All: Train One Network and Specialize it for Efficient DeploymentSaved credit: Contributor · 2020Explore people connected to this work →
- AWQ: Activation-aware Weight Quantization for LLM Compression and AccelerationSaved credit: Contributor · 2024Explore people connected to this work →
- Efficient Streaming Language Models with Attention SinksSaved credit: Contributor · 2024Explore people connected to this work →