张宸 - 上海交通大学

Biography

I joined the School of Electronic Information and Electrical Engineering at Shanghai Jiao Tong University in May 2023 as a Tenure-Track Assistant Professor and Ph.D. advisor. I am also a recipient of the Shanghai Overseas High-Level Talent Program. My research focuses on AI processor architectures and chip systems.

I received my Ph.D. from Peking University in 2017, under the supervision of Prof. Jason Cong and Prof. Guangyu Sun. From 2015 to 2016, I conducted academic research as a visiting scholar at the University of California, Los Angeles (UCLA). After graduation, I worked at Microsoft Research (as a Senior Researcher) and later at Alibaba T-Head Semiconductor (as a core member of the chief architect team), where I led the development of several high-impact AI processors and systems both in China and internationally.

I have published over 30 papers in top-tier conferences and journals, including ISCA, MICRO, FPGA, DAC, and IEEE TCAD, receiving five Best Paper Awards or nominations (FPGA’15, TCAD’19, MICRO’22, ISEDA’25, ISCA’25). My work has been cited over 5,200 times on Google Scholar, with a single paper cited more than 2,600 times. I have been honored as a member of the FPGA and Reconfigurable Computing Hall of Fame, named an AI 2000 Most Influential Scholar in the World, recognized as a Top 2% Most-Cited Researcher by Stanford and Elsevier, and received the ACM ChinaSys Rising Star Award, among other accolades.

News

  • [2025 May.] “DATIS: DRAM Architecture and Technology Integrated Simulation” published on ISEDA-2025 wins Best Paper Award!【Details
  • [2025 Mar.] “Optimizing FPGA-based accelerator design for deep convolutional neural networks” is inducted into the Class of 2025 FPGA and Reconfigurable Computing Hall of Fame!【2017年开始由TCFPGA组委会投票遴选,首次亚太地区第一单位获奖,UCLA NewsSIC NewsSJTU NewsSJTU News
  • [2022 Oct.] I am honored with ChinaSys Rising Star Award. 【ChinaSys Web
  • [2019 Jan.] “Caffeine: Toward uniformed representation and acceleration for deep convolutional neural networks” published on T-CAD 2018 has win Donald O. Pederson Best Paper! 【1999年开始,由IEEE CEDA委员会投票遴选,首次中国第一单位获奖,UCLA NewsPKU News

Selected Publications(Full List

  • H^2-LLM: Hardware-Dataflow Co-Exploration for Heterogeneous Hybrid-Bonding-based Low-Batch LLM Inference, ISCA, 2025
  • SynGPU: Synergizing CUDA and Bit-Serial Tensor Cores for Vision Transformer Acceleration on GPU, DAC, 2025
  • Oltron: Software-Hardware Co-design for Outlier-Aware Quantization of LLMs with Inter-/Intra-Layer Adaptation, DAC, 2024
  • Dual-side sparse tensor core, ISCA, 2021
  • Caffeine: Toward uniformed representation and acceleration for deep convolutional neural networks, T-CAD, 2018
  • Optimizing FPGA-based accelerator design for deep convolutional neural networks, FPGA, 2015

Awards and Honors

  • [2025] FPGA and Reconfigurable Computing Hall of Fame (名人堂)【2017年开始,由TCFPGA组委会投票遴选,首次亚太地区第一单位获奖,SJTU NewsUCLA NewsTCFPGA Web
  • [2024] Stanford and Elsevier Top-2% Most Cited Scholars (Computer Architecture and Hardware)【Web
  • [2021~2024] AI-2000 World’s Most Influential Scholars (AI Chip)【AI-2000 Web
  • [2024] FPGA会议历史(自1992年以来)高被引论文(第一名)【ACM Library
  • [2022] MICRO Top Picks Honorable Mention (MICRO 年度最佳论文优胜奖)
  • [2022] ACM ChinaSys Rising Star【Web
  • [2019] Microsoft Special Stock Award(微软研究院院长特别奖)
  • [2019] Donald O. Pederson Best Paper 【1999年开始,由IEEE CEDA委员会投票遴选,首次中国第一单位获奖,IEEE NewsUCLA NewsPKU News
  • [2019] Microsoft Research Special Stock Award
  • [2015] FPGA Best Paper Nomination 【Web