(2023). TAP: Efficient Derivation of Tensor Parallel Plans for Large Neural Networks. In ISCA'23 ASSYST Workshop.

PDF Cite Poster

(2023). ParaGAN: A Cloud Training Framework for Generative Adversarial Networks. In ISCA'23 MLArchSys Workshop.

PDF Cite Poster

(2022). Whale: Efficient Giant Model Training over Heterogeneous GPUs. In USENIX ATC'22.

PDF Cite Code Slides

(2021). Going Wider Instead of Deeper. In AAAI'22.

PDF Cite