基本信息
导师姓名：陈凯
担任职务：北京中关村学院 导师
主要研究领域：包括大模型、生成式AI、科学智能、多模态等，曾任微软亚洲研究院首席研究员，研发的人工智能关键技术在Windows、Office、Azure等微软重要产品落地，中国科学技术大学少年班学院学士，中国科大-微软亚洲研究院联合培养博士。
个人经历
教育经历：
2007.8 – 2011.6 中国科学技术大学 少年班学院
2011.9 – 2016.6 中国科学技术大学-微软亚洲研究院 联合培养博士
工作经历：
2023.8-2024.12 北京智源人工智能研究院
2023.3-2023.8 独立开发者
2016.6-2023.3 微软亚洲研究院
科学研究
代表性学术论文：
l Yunfei Teng, Yuxuan Ren, Kai Chen*, Xi Chen, Zhaoming Chen, Qiwei Ye, “CryoGEN: Cryogenic Electron Tomography Reconstruction via Generative Energy-base Models”, ICLR-2025
l Jingxuan Feng, Lili Wang, Xiaoya Zhai, Kai Chen, Wenming Wu, Ligang Liu, Xiao-Ming Fu, “Constructing boundary-identical microstructures via guided diffusion for fast multiscale topology optimization”, Computer Methods in Applied Mechanics and Engineering 436, 117735
l Yukang Yang, Dongnan Gui, Yuhui Yuan, Haisong Ding, Han Hu, Kai Chen, “GlyphControl: Glyph Conditional Control for Visual Text Generation”, NeurIPS-2024
l Dongnan Gui, Kai Chen*, Haisong Ding, Qiang Huo,“Zero-shot Generation of Training Data with Denoising Diffusion Probabilistic Model for Handwritten Chinese Character Recognition”, ICDAR-2023
l Haisong Ding, Bozhi Luan, Dongnan Gui, Kai Chen*, Qiang Huo, “Improving Handwritten OCR with Training Samples Generated by Glyph Conditional Denoising Diffusion Probabilistic Model”, ICDAR-2023
l Haisong Ding, Kai Chen*, Qiang Huo, “An Encoder-Decoder Approach to Handwritten Mathematical Expression Recognition with Multi-head Attention and Stacked Decoder”, in Proc. ICDAR-2021,pp.602-616
l Mingyang Guan, Haisong Ding, Kai Chen*, Qiang Huo, “Improving Handwritten OCR with Augmented Text Line Images Synthesized from Online Handwriting Samples by Style Conditioned GAN”, in Proc. ICFHR-2020, pp.151-156 (Best Student Paper).
l Haisong Ding, Kai Chen*, Qiang Huo, “Improving Knowledge Distillation of CTC-Trained Acoustic Models with Alignment-Consistent Ensemble and Target Delay”, in IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 28, pp. 2561-2571, 2020.
l Kai Chen, Haisong Ding, Qiang Huo, “Parallelizing Adam with Blockwise Model-Update Filtering”, in Proc. ICASSP-2020, pp.3027-3031
l Haisong Ding, Kai Chen*, Qiang Huo, “Compression of CTC-Trained Acoustic Models by Dynamic Frame-Wise Distillation or Segment-Wise N-Best Hypotheses Imitation”, in Proc. INTERSPEECH-2019, pp.3218-3222.
l Haisong Ding, Kai Chen*, Qiang Huo, “Compressing CNN-DBLSTM models for OCR with teacher-student learning and Tucker decomposition”, in Pattern Recognition, 2019, vol.96, 106957.
l Haisong Ding, Kai Chen*, Wenping Hu, Meng Cai, Qiang Huo, “Building Compact CNNDBLSTM Based Character Models for Handwriting Recognition and OCR by Teacher Student Learning”, in Proc. ICFHR-2018, pp.139-144.
l Kai Chen, Li Tian, Haisong Ding, Meng Cai, Lei Sun, Sen Liang, Qiang Huo, “A Compact CNNDBLSTM Based Character Model for Online Handwritten Chinese Text Recognition”, in Proc.ICDAR-2017, pp.1068-1073.
l Haisong Ding, Kai Chen*, Ye Yuan, Meng Cai, Lei Sun, Sen Liang, Qiang Huo, “A Compact CNN-DBLSTM based character model for Offline Handwriting Recognition with Tucker Decomposition”,in Proc. ICDAR-2017, pp.507-512.
l Meng Cai, Wenping Hu, Kai Chen, Lei Sun, Sen Liang, XiongJian Mo, Qiang Huo, “An open vocabulary OCR system with hybrid word-subword language models”, in ICDAR-2017, pp.519- 524
l Kai Chen, Qiang Huo, “Training Deep Bidirectional LSTM Acoustic Model for LVCSR by a Context-Sensitive-Chunk BPTT Approach”, in IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol.24, pp.1185-1193, 2016
l Kai Chen, Zhi-Jie Yan, Qiang Huo, “Training Deep Bidirectional LSTM Acoustic Model for LVCSR by a Context-Sensitive-Chunk BPTT Approach” in INTERSPEECH-2015, pp.3600-3604
l Wenping Hu, Meng Cai, Kai Chen, Haisong Ding, Lei Sun, Sen Liang, Xiongjian Mo, and Qiang Huo, “Sequence discriminative training for offline handwriting recognition by an interpolated CTC and lattice-free MMI objective function”, in Proc. ICDAR-2017, pp.519-524.
l Kai Chen, Qiang Huo, “Scalable Training of Deep Learning Machines by Incremental Block Training with Intra-Block Parallel Optimization and Blockwise Model-Update Filtering”, in Proc. ICASSP-2016, pp.5880-5884.
l Kai Chen, Zhi-Jie Yan, Qiang Huo, “A Context-Sensitive-Chunk BPTT Approach to Training Deep LSTM/BLSTM Recurrent Neural Networks for Offline Handwriting Recognition”, in Proc. ICDAR-2015, pp.411-415.
l Jun Du, Qiang Huo, Kai Chen, “Designing Compact Classifiers for Rotation-Free Recognition of Large Vocabulary Online Handwritten Chinese Characters”, in Proc. ICASSP-2012, pp.1721- 1724.