Scholar
Xiaofan Zhang
Google Scholar ID: gG24R6MAAAAJ
Google, UIUC
Citations & Impact
Citations
5,305
H-index
23
i10-index
31
Publications
20
Co-authors
13
list available
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- 2025: ASAP: an Agentic Solution to Auto-optimize Performance of Large-Scale LLM Training (NeurIPS ML for Systems Workshop)
- 2025: Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities (arXiv preprint arXiv:2507.06261)
- 2025: Reconfigurable Stream Network Architecture (ISCA)
- 2025: Profile-Guided Quantization: a compiler solution to automate quantization for efficient LLM training (ISCA MLArchSys workshop)
- 2025: SSDTrain: An Activation Offloading Framework to SSDs for Faster Large Language Model Training (DAC)
- 2024: ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization (NeurIPS)
- 2024: New Solutions on LLM Acceleration, Optimization, and Application (DAC)
- 2024: AutoAI2C: An Automated Hardware Generator for DNN Acceleration on both FPGA and ASIC (IEEE Transactions on Computer-Aided Design)
Research Experience
- Staff Software Engineer at Google, working on large-scale AI systems to enable efficient Gemini training and serving on TPUs.
Education
- Received Ph.D. from the University of Illinois Urbana-Champaign (UIUC) in 2022, supervised by Prof. Deming Chen, and collaborated closely with Prof. Wen-mei Hwu and Prof. Junjun Xiong; B.S. and M.S. from UESTC in Chengdu, China.
Background
- AI Systems Engineer & Researcher, with research interests in AI Systems, Energy-efficient Computing, and Hardware/Software Co-design.
Miscellany
- Contact: xiaofanz [at] google
- Google Scholar and LinkedIn profiles.
