Xu Sun 孙栩   Associate Professor, PhD Supervisor Director of Language Computing and Machine Learning Group, Department of Computer Science, Peking University Email：xusun (AT) pku.edu.cn       Github     Google Scholar     Brief Bio: Xu Sun is an associate professor, PI and director of the Language Computing and Machine Learning Group of ICL in School of Computer Science, Peking University. He got Ph.D. of CS from The University of Tokyo (2010), advised by Prof. Jun'ichi Tsujii, and M.S. of CS from Peking University (2007). From 2010 to 2012, he worked at The University of Tokyo, Cornell University, and The Hong Kong Polytechnic University as research fellows. He has been a research intern at MSR-Redmond in 2009. His research focuses on natural language processing and machine learning, especially on natural language generation, multi-modal NLP, and AIGC. He received the COLING 2018 Best Paper Award and the EMNLP 2023 Best Long Paper Award.

中文简介

Our Research Group

• Language Computing and Machine Learning Group

Research Interests

• Natural language generation, AIGC (large language models, video captioning, long video understanding, etc.)


• Multi-modal NLP (vision-language processing, text-to-video generation, etc.)


• Deep learning (in-context learning, prompt learning, deep neural networks, etc.)

Our Github Tools

• AdaBound (An optimizer that trains as fast as Adam and as good as SGD, ICLR 2019)
• pkuseg (A multi-domain Chinese word segmentation toolkit, arXiv 2019)

Publications

• Full paper list


• Selected papers


• TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding. S.Ren, L.Yao, S.Li, X.Sun, L.Hou
  In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
  
  
• Towards Codable Text Watermarking for Large Language Models. L.Wang, W.Yang, D.Chen, H.Zhou, Y.Lin, F.Meng, J.Zhou, X.Sun
  In Proceedings of the Twelfth International Conference on Learning Representations (ICLR), 2024
  
  
• Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning. L.Wang, L.Li, D.Dai, D.Chen, H.Zhou, F.Meng, J.Zhou, X.Sun
  In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023 [Best Long Paper Award]
  
  
• Fed-FA: Theoretically Modeling Client Data Divergence for Federated Language Backdoor Defense. Z.Zhang, D.Chen, H.Zhou, F.Meng, J.Zhou, X.Sun
  In Proceedings of the Thirty-seventh Annual Conference on Neural Information Processing Systems (NeurIPS), 2023
  
  
• Retrieve, Reason, and Refine: Generating Accurate and Faithful Patient Instructions. F.Liu, B.Yang, C.You, X.Wu, S.Ge, Z.Liu, X.Sun, Y.Yang, D.A.Clifton
  In Proceedings of the Thirty-sixth Annual Conference on Neural Information Processing Systems (NeurIPS), 2022
  
  
• How to Inject Backdoors with Better Consistency: Logit Anchoring on Clean Data. Z.Zhang, L.Lyu, W.Wang, L.Sun, X.Sun
  In Proceedings of the International Conference on Learning Representations (ICLR), 2022
  
  
• Aligning Source Visual and Target Language Domains for Unpaired Video Captioning. F.Liu, X.Wu, C.You, S.Ge, Y.Zou, X.Sun
  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2022
  
  
• Topology-Imbalance Learning for Semi-Supervised Node Classification. D.Chen, Y.Lin, G.Zhao, X.Ren, P.Li, J.Zhou, X.Sun
  In Proceedings of the Thirty-fifth Annual Conference on Neural Information Processing Systems (NeurIPS), 2021
  
  
• Auto-Encoding Knowledge Graph for Unsupervised Medical Report Generation. F.Liu, C.You, X.Wu, S.Ge, S.Wang, X.Sun
  In Proceedings of the Thirty-fifth Annual Conference on Neural Information Processing Systems (NeurIPS), 2021
  
  
• Prophet Attention: Predicting Attention with Future Attention for Improved Image Captioning. F.Liu, X.Ren, X.Wu, S.Ge, W.Fan, Y.Zou, X.Sun
  In Proceedings of the Thirty-fourth Annual Conference on Neural Information Processing Systems (NeurIPS), 2020
  
  
• Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological View. D.Chen, Y.Lin, W.Li, P.Li, J.Zhou, X.Sun
  The Thirty-fourth AAAI Conference on Artificial Intelligence (AAAI), 2020
  
  
• Understanding and Improving Layer Normalization. J.Xu, X.Sun, Z.Zhang, G.Zhao, J.Lin.
  In Proceedings of the Thirty-third Annual Conference on Neural Information Processing Systems (NeurIPS), 2019
  
  
• Aligning Visual Regions and Textual Concepts for Semantic-Grounded Image Representations. F.Liu^#, Y.Liu^#, X.Ren^#, X.He, K.Lei, X.Sun.
  In Proceedings of the Thirty-third Annual Conference on Neural Information Processing Systems (NeurIPS), 2019
  
  
• Adaptive Gradient Methods with Dynamic Bound of Learning Rate. L.Luo^#, Y.Xiong^#, Y.Liu, X.Sun.
  In Proceedings of the International Conference on Learning Representations (ICLR), 2019
  
  
• Global Encoding for Abstractive Summarization. J.Lin, X.Sun, S.Ma, Q.Su.
  In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), 2018
  
  
• Unpaired Sentiment-to-Sentiment Translation: A Cycled Reinforcement Learning Approach. J.Xu, X.Sun,Q.Zeng, X.Zhang, X.Ren, H.Wang, W.Li.
  In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), 2018
  
  
• DP-GAN: A Diversity-Promoting Generative Adversarial Network for Generating Informative and Diversified Text. J.Xu, X.Ren, J.Lin, X.Sun.
  In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2018
  
  
• SGM: Sequence Generation Model for Multi-label Classification. P.Yang, X.Sun, W.Li, S.Ma, W.Wu, H.Wang.
  The 27th International Conference on Computational Linguistics (COLING), 2018 [Best Paper Award]
  
  
• meProp: Sparsified Back Propagation for Accelerated Deep Learning with Reduced Overfitting. X.Sun, X.Ren, S.Ma, H.Wang.
  The Thirty-fourth International Conference on Machine Learning (ICML), 2017
  
  

Awards

• EMNLP 2023 Best Long Paper Award, 2023


• FinNLP Workshop at IJCAI 2022 Best Paper Award, 2022


• Young Scientist, Beijing Academy of Artifical Intelligence, 2020 (北京智源青年科学家)


• COLING 2018 Best Paper Award, 2018


• 1st Rank Prize, Science and Technology Prize of CIE (Chinese Institute of Electronics), 2018 (中国电子学会科学技术奖一等奖)


• Architecture Innovation Award, JD Dialogue Challenge, 2018


• CCF NLPCC Distinguished Young Scientist Award, 2018


• Okawa Research Award, Japan, 2017


• Boya Young Fellow, Peking University, 2016 (北京大学博雅青年学者)


• Qiu Shi Outstanding Young Scholar Award (求是杰出青年学者奖), Qiu Shi Foundation, 2015

Academic Activities

• Senior Area Chair(SAC) / Area Chair(AC) / Senior Program Committee (SPC): ACL 2021(SAC), EMNLP 2020(AC), EMNLP 2015(AC), AAAI 2020(SPC), IJCAI 2020(SPC), IJCAI 2018(SPC), NLPCC 2018(AC), IJCNLP 2017(AC), YCCL 2012(AC)


• Program committee member: ACL, ICML, NIPS, IJCAI, AAAI, EMNLP, COLING, NAACL, ACML, PAKDD, CCL, NLPCC, LREC


• Journal reviewer: IEEE TPAMI, IEEE TNNLS, IEEE TKDE, Computational Linguistics, TACL, IEEE TASLP, Information Processing and Management, ACM TALIP, KAIS

Work Experience

• Tenured Associate Professor and PhD Supervisor, Peking University, 2019-now


• Tenure-Track Faculty and PhD Supervisor, Peking University, 2012-2018


• Researcher, The Hong Kong Polytechnic University, 2012


• Researcher, Cornell University, 2011


• Researcher, The University of Tokyo, 2010 - 2011

Students