获得国家人才项目支持，CCF开源发展委员会执委委员，中国医疗保健国际交流促进会健康数据与数字医学分会委员。2018.6—2019.1公派到UIUC的Laxmikant V. Kale教授课题组访学。长期从事大模型、图计算、智能工业软件和数值反应堆等的研究；近几年智超融合领域的研究成果在能源、材料、气象等多个学科领域进行成果应用和落地。在国内外高水平期刊和会议IEEE TPDS、IEEE TSE、IEEE FS、Innovation、PPoPP、SC、HPDC、DAC、AAAI、CPC、中国科学、计算机学报和软件学报等发表学术论文100余篇；授权专利60余项；专著2部。获得SC 23（CCF A推荐）最佳论文和最佳学生论文双提名，KSEM 22最佳学生论文。获得北京市技术发明等多个省部级奖、中核集团技术发明一等奖（2次）和二等奖（1次）、国家电网有限公司科技进步三等奖等、2025 年度中国高影响力数据论文——工程技术领域。先后承担了自然科学基金、863子课题、国家重点研发计划课题、中国科学院战略性先导A科技专项课题、国家电网基础性/前瞻性项目（院士合作项目），国家电网公司总部科技项目等。

[1] ANT-MOC: Scalable Neutral Particle Transport Using 3D Method of Characteristics on Multi-GPU Systems，SC '23: International Conference for High Performance Computing, Networking, Storage and Analysis (唯一通信作者，最佳论文和最佳学生论文双提名，CCF A)

[2] The numerical reactor system for China’s new-generation exascale supercomputer, The Innovation, 2026（唯一通信作者，Cell 子刊）

[3] HIP-DFPT: Scalable Optimization of Irregular Workloads in Quantum Perturbation on GPU Clusters, IEEE Transactions on Parallel and Distributed Systems, 2026 （唯一通信作者，CCF A）

[4] TAC: Cache-Based System for Accelerating Billion-Scale GNN Training on Multi-GPU Platform, PPoPP '26: the ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming (第一通信作者，CCF A)

[5] Full-Core Fluid-Structure-Interaction Simulation of Nuclear Reactor on CPU+GPU Hybrid Clusters, HPDC’ 26: the International ACM Symposium on High-Performance Parallel and Distributed Computing (唯一通信作者，CCF A)

[6] PillarSparse: Rethinking Unstructured Sparse Formats for Tensor Cores. DAC’ 26: the 63st ACM/IEEE Design Automation Conference  (唯一通信作者，CCF A)

[7] Acc-SpMV: Accelerating General-purpose Sparse Matrix-Vector Multiplication with GPU CUDA Cores. DAC’ 26: the 63st ACM/IEEE Design Automation Conference  (CCF A)

[8] RipAlert: A Future-Frame-Aware Framework for Rip Current Forecasting and Early Alerting. AAAI' 26: the 40th AAAI Conference on Artificial Intelligence (唯一通信作者，CCF A)

[9] PhysTrans: A Physics-Aware Transferable Framework for Global Cold-Start Photovoltaic Forecasting. IJCAI’ 26: the 35th International Joint Conference on Artificial Intelligence (唯一通信作者，CCF B)

[10] NeuroDALEC: A Differentiable and Interpretable Mass-Conserving Framework for Terrestrial Ecosystem Carbon Cycle Dynamics. IJCAI’ 26: the 35th International Joint Conference on Artificial Intelligence (唯一通信作者，CCF B)

[11] AI for science: Progress, challenges, and perspectives, The Innovation, 2026 (Cell 子刊)

[12] Understanding as compression: A new evaluation framework for large language models, The Innovation Informatics, 2025 （唯一通信作者，Cell 子刊）

[13] ParGNN: A Scalable Graph Neural Network Training Framework on multi-GPUs, DAC' 25：ACM/IEEE Design Automation Conference (唯一通信作者，CCF A)

[14] MISA-AKMC: Achieve Kinetic Monte Carlo Simulation of 20 Quadrillion Atoms on GPU Clusters, SC '25: the International Conference for High Performance Computing, Networking, Storage and Analysis (唯一通信作者，CCF A)

[15] Acc-SpMM: Accelerating General-purpose Sparse Matrix-Matrix Multiplication with GPU Tensor Cores, PPoPP' 25: the ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming (唯一通信作者，CCF A)

[16] SEP: A General Lossless Compression Framework with Semantics Enhancement and Multi-Stream Pipelines, IJCAI’ 25: the 24st International Joint Conference on Artificial Intelligence (唯一通信作者，CCF A)

[17] MCloudNet: An Ultra-Short-Term Photovoltaic Power Forecasting Framework With Multi-Layer Cloud Coverage, IJCAI’ 25: the 24st International Joint Conference on Artificial Intelligence (唯一通信作者，CCF A)

[18] POSTER: ParGNN: Efficient Training for Large-Scale Graph Neural Network on GPU Clusters, PPoPP’ 24: the ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming  (唯一通信作者，CCF A)

[19] InParformer: Evolutionary Decomposition Transformers with Interactive Parallel Attention for Long-Term Time Series Forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence; AAAI’23 (唯一通信作者，CCF A)

[20] Large-Scale Simulation of Structural Dynamics Computing on GPU Clusters, SC' 23: International Conference for High Performance Computing, Networking, Storage and Analysis (唯一通信作者，CCF A)

[21]  A Scalable Hybrid TotalFETI Method for Massively Parallel FEM Simulations, PPoPP’23: ACM SIGPLAN Annual Symposium Principles and Practice of Parallel Programming (唯一通信作者，CCF A)

[22] Day-ahead PV power forecasting model based on fine-grained temporal attention and cloud-coverage spatial attention, IEEE Transactions on Sustainable Energy,2023 (唯一通信作者)

[23] Intra-hour Photovoltaic Generation Forecasting based on Multi-source Data and Deep Learning Methods, IEEE Transactions on Sustainable Energy, 2022 (唯一通信作者)

[24] A Multi-level Attention-based LSTM Network for Ultra-short-term Solar Power Forecast using Meteorological Knowledge. KSEM’ 22: the 15th International Conference on Knowledge Science, Engineering and Management. (唯一通信作者, CCF推荐), 最佳学生论文（3/498）