Dongcheng Zhao

Dongcheng Zhao

Associate Professor

I am an Associate Professor at the CAS Center for Excellence in Brain Science and Intelligence Technology (CEBSIT).

My research focuses on biologically plausible and computationally efficient spiking neural networks (SNNs), aiming to bridge principles from neuroscience with scalable machine intelligence.

My research has been published in leading journals and conferences, including Science Advances, PNAS, Patterns (Cell Press), iScience, Scientific Data (Nature Portfolio), IEEE Transactions, as well as top-tier AI venues such as ICLR, NeurIPS, ICML, CVPR, ICCV.

I am currently recruiting highly motivated interns with strong interest in brain-inspired intelligence and spiking neural networks.

🔥 News

📝 Selected Publications

Inverse effectiveness driven multimodal fusion: Incorporating brain-inspired mechanisms for multimodal learning in artificial intelligence
Science Advances, 2026
Xiang He, Dongcheng Zhao, Yang Li, Qingqun Kong, Xin Yang, Yi Zeng (equal contribution)

NeuEvo: Brain-inspired Neural Circuit Evolution for Spiking Neural Networks
Proceedings of the National Academy of Sciences (PNAS), 2023
Guobin Shen, Dongcheng Zhao, Yiting Dong, Yi Zeng (equal contribution)

ETC: Improving Stability and Performance of Spiking Neural Networks through Enhancing Temporal Consistency
Pattern Recognition, 2025
Dongcheng Zhao, Guobin Shen, Yiting Dong, Yang Li, Yi Zeng

➡️ Complete publication list

🧩 Grants & Projects

2025.01–2027.12
国家自然科学基金委员会 · 青年科学基金 · 主持
融合神经元精细建模的多神经环路以及学习机制协同的脉冲神经网络研究

2025.07–2026.07
企业横向项目 · 主持
大模型对齐机制研究

2024.01–2027.12
国家自然科学基金委员会 · 面上项目 · 参与
基于大脑组织结构和神经回放多样性的连续学习方法研究

2024.09–2026.09
北京市科学技术委员会 · 中央引导地方专项 · 参与
基于外部监督对齐与内部机制对齐的超级对齐关键技术研究及示范应用

🚀 Selected Projects

PandaGuard

PandaGuard structures LLM safety evaluation as an interactive system with Attackers, Defenders, and Judges.
It supports plug-and-play algorithms, multiple inference backends, and flexible human-in-the-loop interfaces.

Project Page · Paper

Spiking Transformer Evaluation Platform (STEP)

STEP provides a unified and reproducible benchmarking pipeline for Spiking Transformers, supporting classification, segmentation, and detection with energy modeling.

Project Page · Paper

Brain-inspired Cognitive Intelligence Engine (BrainCog) BrainCog is an open-source SNN-based cognitive intelligence engine for brain-inspired AI, embodied intelligence, and brain simulation.

Project Page · Paper

🎖 Honors and Awards

  • Cell Press China Paper of the Year Award (Interdisciplinary Science), 2023
  • Cell Press China Paper of the Year Award (Interdisciplinary Science), 2022

📖 Education

  • Ph.D., Institute of Automation, Chinese Academy of Sciences (CAS)
  • B.Sc., School of Mathematics and Statistics, Xidian University

💻 Academic Service

  • Program Committee / Reviewer: IJCAI, AAAI, ECCV, ICCV, CVPR, ICML, NeurIPS, ICML, ICLR
  • Journal Reviewer: IEEE TPAMI, TNNLS, TMM, TIP, TETCI

📬 Contact