About Me
Dr. Juntao Li is now an associate professor at School of Computer Science & Technology, Soochow University, working with Prof. Min Zhang. Before that, he obtained a doctoral degree from Peking University in 2020, supervised by Dr. Rui Yan, Prof. Dongyan Zhao, and Prof. Dongmin Chen.
I am leading a research group on language models. We are dedicated to building application-oriented, open-source, large-size language models with transparency, reusability, and low-cost deployment. More details can be found at OpenNLG.
1. Contact
ljtsudanlp@gmail.com
ljt@suda.edu.cn
2. Research Interests and Selected Papers
(‘*’ = equal contribution, ’#’ = corresponding author )
2.1 Pretrained Language Models
- (OpenBA: An Open-sourced 15B Bilingual Asymmetric seq2seq Model Pre-trained from Scratch)[paper][Model & Code], SCIS, CCF-A
- [COLM-2024]Timo: Towards Better Temporal Reasoning for Language Models
- [ACL-24] Feature-based Low-Rank Compression of Large Language Models via Bayesian Optimization.
- [ACL-24] Exploring Reversal Mathematical Reasoning Ability for Large Language Models.
- [ACL-24] Living in the Moment: Can Large Language Models Grasp Co-Temporal Reasoning?.
- [ACL-24] CMD: a framework for Context-aware Model self-Detoxification.
- [ACL-24] Demonstration Augmentation for Zero-shot In-context Learning.
- [ACL-24] Rethinking Negative Instances for Generative Named Entity Recognition.
- [ICML-24] Zecheng Tang, Chenfei Wu, Zekai Zhang, Mingheng Ni, Shengming Yin, Yu Liu, Zhengyuan Yang, Lijuan Wang, Zicheng Liu, Juntao Li, Nan Duan. StrokeNUWA: Tokenizing Strokes for Vector Graphic Synthesis.
- [ICLR-24] Yisheng Xiao, Juntao Li#, Zechen Sun, Zechang Li, Qingrong Xia, Xinyu Duan, Zhefeng Wang, Min Zhang. Are Bert Family Good Instruction Followers? A Study on Their Potential And Limitations.
- [ICLR-24] Zecheng Tang, Chenfei Wu, Juntao Li, Nan Duan. LayoutNUWA: Revealing the Hidden Layout Expertise of Large Language Models.
- [EMNLP-23] Yi Su, Yixin Ji, Juntao Li#, Hai Ye, Min Zhang. Beware of Model Collapse! Fast and Stable Test-time Adaptation for Robust Question Answering. EMNLP 2023. [pdf](CCF B)
- [EMNLP-23] Chuyue Zhou, WangJie You, Juntao Li#, Jing Ye, Kehai Chen, Min Zhang. INFORM : Information eNtropy based multi-step reasoning FOR large language Models. EMNLP 2023. [pdf](CCF B)
- [EMNLP-23] Yixin Ji, Jikai Wang, Juntao Li#, Hai Ye, Min Zhang. Isotropic Representation Can Improve Zero-Shot Cross-Lingual Transfer on Multilingual Language Models. EMNLP 2023 Findings. [pdf]
- [EMNLP-23] Zhaochen Su, Juntao Li#, Zikang Zhang, Zihan Zhou, Min Zhang . Efficient Continue Training of Temporal Language Model with Structural Information. EMNLP 2023 Findings. [pdf]
- [EMNLP-23] Haoke Zhang, Yue Wang, Juntao Li#, Xiabing Zhou, Min Zhang. G-SPEED: General SParse Efficient Editing MoDel. EMNLP 2023 Findings. [pdf]
- [EMNLP-23] Pei Guo, yisheng xiao, Juntao Li#, Yixin Ji, Min Zhang. Isotropy-Enhanced Conditional Masked Language Models. EMNLP 2023 Findings. [pdf]
- [EMNLP-23] Lei Geng, Xu Yan, Ziqiang Cao, Juntao Li, Wenjie Li, Sujian Li, Xinjie Zhou, Yang Yang, Jun Zhang. KBioXLM: A Knowledge-anchored Biomedical Multilingual Pretrained Language Model. EMNLP 2023 Findings. [pdf]
- [AI J] Yue Wang, Lijun Wu, Juntao Li#, Xiaobo Liang, Min Zhang. Are BERT Families Zero-Shot Learners? A Study on Their Potential and Limitations. Artificial Intelligence. [pdf][code](CCF A)
- [ACL-23] Yixin Ji, Jikai Wang, Juntao Li#, Qiang Chen, Wenliang Chen and Min Zhang. Early Exit with Disentangled Representation and Equiangular Tight Frame. [pdf][code] ACL 2023 Findings
- [EMNLP-22] Zhaochen Su*, Zecheng Tang*, Xinyan Guan, Juntao Li#, Lijun Wu, Min Zhang. Improving Temporal Generalization of Pre-trained Language Models with Lexical Semantic Change. [pdf][code] In EMNLP’22. (CCF B)
- [COLING-22] Dan Qiao, Chenchen Dai, Yuyang Ding, Juntao Li#, Qiang Chen, Wenliang Chen and Min Zhang. SelfMix: Robust Learning Against Textual Label Noise with Self-Mixup Training. In COLING’22. (Oral) [pdf][code] (CCF B)
- [EMNLP-20] Hai Ye, Qingyu Tan, Ruidan He, Juntao Li, Hwee Tou Ng, Lidong Bing. Feature Adaptation of Pre-Trained Language Models across Languages and Domains with Robust Self-Training. In EMNLP’20. Full paper. [pdf][code] (CCF B)
- [IJCAI-20] Juntao Li, Ruidan He, Hai Ye, Hwee Tou Ng, Lidong Bing, and Rui Yan. Unsupervised Domain Adaptation of a Pretrained Cross-Lingual Language Model. In IJCAI-PRICAI’20. Full paper. [pdf] [code](CCF A)
2.2 Natural Language Generation
- [ACL-24] Efficient Domain Adaptation for Non-Autoregressive Machine Translation.
- [TPAMI] Juntao Li, Xiaobo Liang, Lijun Wu, Yue Wang, Qi Meng, Tao Qin, Tie-yan Liu, Min Zhang. Randomness Regularization with Simple Consistency Training for Neural Networks. TPAMI, 2024
- [TPAMI] Yisheng Xiao, Lijun Wu, Junliang Guo, Juntao Li#, Min Zhang, Tao Qin, Tie-yan Liu. A Survey on Non-Autoregressive Generation for Neural Machine Translation and Beyond. TPAMI, 2023. [pdf] (CCF A)
- [WSDM-24] Yue Wang, Zilong Zheng, Zecheng Tang, Juntao Li#, Zhihui Liu, Kunlong Chen, Jinxiong Chang, Qishen Zhang, Zhongyi Liu, Min Zhang. Towards Better Chinese Spelling Check for Search Engines: A New Dataset and Strong Baseline. WSDM 2024. [pdf](CCF B)
- [NeurIPS-23] Tong Wu, Zhihao Fan, Xiao Liu, Yeyun Gong, Yelong Shen, Jian Jiao, Hai-Tao Zheng, Juntao Li, Zhongyu Wei, Jian Guo, Nan Duan, Weizhu Chen. AR-Diffusion: Auto-Regressive Diffusion Model for Text Generation. WSDM 2024. [pdf](CCF A)
- [ACL-23] Xiaobo Liang, Zecheng Tang, Juntao Li#, Min Zhang. Open-ended Long Text Generation via Masked Language Modeling. [pdf][code] ACL 2023 (CCF A)
- [ACL-23] Xiaobo Liang, Juntao Li#, Lijun Wu, Ziqiang Cao and Min Zhang. Dynamic and Efficient Inference for Text Generation via BERT Family. [pdf][code] ACL 2023 (CCF A)
- [ACL-23] Yue Wang, Dan Qiao, Juntao Li#, Jinxiong Chang, Qishen Zhang, Zhongyi Liu, Guannan Zhang and Min Zhang. Towards Better Hierarchical Text Classification with Data Generation. [pdf][code] ACL 2023 Findings
- [ACL-23] Zecheng Tang, Pinzheng Wang, Keyan Zhou, Juntao Li#, Ziqiang Cao and Min Zhang. Can Diffusion Model Achieve Better Performance in Text Generation ? Bridging the Gap between Training and Inference! [pdf][code] ACL 2023 Findings
- [AAAI-23] Pei Guo, Yisheng Xiao, Juntao Li#, Lijun Wu, Min Zhang. RenewNAT: Renewing Potential Translation for Non-Autoregressive Transformer. [pdf][code] In AAAI’23 (CCF A)
- [AAAI-23] Yisheng Xiao, Lijun Wu, Ruiyang Xu, Juntao Li#, Tao Qin, Tie-yan Liu, Min Zhang. AMOM: Adaptive Masking over Masking for Conditional Masked Language Model. [pdf][code] In AAAI’23 (CCF A)
- [NeurIPS-21] Xiaobo Liang, Lijun Wu, Juntao Li#, Yue Wang, Qi Meng, Wei Chen, Tao Qin, Min Zhang, Tie-yan Liu. R-Drop: Regularized Dropout for Neural Networks.In NeurIPS’21. Full paper. [pdf][code] (CCF A)
- [AAAI-21] Meng-Hsuan Yu, Juntao Li*, Zhangming Chan, Dongyan Zhao and Rui Yan. Content Learning with Structure-Aware Writing: A Graph-Infused Dual Conditional Variational Autoencoder for Automatic Storytelling. In AAAI’21. Full paper. [pdf] (CCF A)
- [AAAI-20] Juntao Li, Chang Liu, Jian Wang, Lidong Bing, Hongsong Li, Xiaozhong Liu, Dongyan Zhao and Rui Yan. Cross-Lingual Low-Resource Set-to-Description Retrieval for Global E-Commerce. In AAAI’20. Full paper. [pdf] (CCF A)
- [AAAI-20] Meng-Hsuan Yu, Juntao Li*, Danyang Liu, Bo Tang, Haisong Zhang, Dongyan Zhao and Rui Yan. Draft and Edit: Automatic Storytelling Through Multi-Pass Hierarchical Conditional Variational Autoencoder. In AAAI’20. Full paper. [pdf] (CCF A)
- [AAAI-20] Danyang Liu, Juntao Li*, Meng-Hsuan Yu, Ziming Huang, Gongshen Liu, Dongyan Zhao and Rui Yan. A Character-Centric Neural Model for Automated Story Generation. In AAAI’20. Full paper. [pdf] (CCF A)
- [AAAI-19] Juntao Li, Lidong Bing, Lisong Qiu, Dongmin Chen, Dongyan Zhao and Rui Yan. Learning to Write Creative Stories with Thematic Consistency. In AAAI’19. Full paper. [pdf] (CCF A)
- [EMNLP-18] Juntao Li, Yan Song, Haisong Zhang, Dongmin Chen, Shuming Shi, Dongyan Zhao, and Rui Yan. Generating Classical Chinese Poems via Conditional Variational Autoencoder and Adversarial Training. In EMNLP’18. Full paper. [pdf] (CCF B)
2.3 Dialogue Systems
- [FnTIR] Rui Yan, Juntao Li#, Zhou Yu. Deep Learning for Dialogue System: Chit-Chat and Beyond. Foundations and Trends® in Information Retrieval. 2022, [pdf] (SCI Q1, IF=8.0, 129 pages + 42 pages references)
- [ACL-23] Chongyang Tao, Jiazhan Feng, Tao Shen, Chang Liu, Juntao Li, Xiubo Geng, Daxin Jiang. CORE: Cooperative Training of Retriever-Reranker for Effective Dialogue Response Selection. [pdf] ACL 2023 (CCF A)
- [TOIS-21] Juntao Li, Chang Liu, Chongyang Tao, Zhangming Chan, DongyanZhao, Min Zhang, Rui Yan. Dialog History Matters! Personalized Response Selection in Multi-turn Retrieval-based Chatbots. TOIS,2021 [pdf] (CCF A)
- [ACL-19] Lisong Qiu, Juntao Li, Wei Bi, Dongyan Zhao and Rui Yan. Are Training Samples Correlated? Learning to Generate Dialogue Responses with Multiple References. In ACL’19. Full paper. [pdf] (CCF A)
- [AAAI-19] Juntao Li, Lisong Qiu, Bo Tang, Dongmin Chen, Dongyan Zhao and Rui Yan. Insufficient Data Can Also Rock! Learning to Converse Using Smaller Data with Augmentation. In AAAI’19. Full paper. [pdf] (CCF A)
- [EMNLP-19] Zhangming Chan, Juntao Li*, Xiaopeng Yang, Xiuying Chen, Wenpeng Hu, Dongyan Zhao, and Rui Yan. Modeling Personalization in Continuous Space for Response Generation via Augmented Wasserstein Autoencoders. In EMNLP’19. Full paper. [pdf] (CCF B)
3. Grants
- 第九届中国科协青年人才托举工程项目, PI
- National Science Foundation of China, No. 62206194, PI
- Natural Science Foundation of Jiangsu Province, No. BK20220488, PI
- Alibaba Innovation Research Grant, Co-PI
- Wudao Open Fund (BAAI), PI
4. Experiences
- Microsoft Research Aisa (2023/03-2023/06), Supervised by Nan Duan
- National University of Singapore (2019/09-2020/02), Supervised by Hwee Tou Ng
- Alibaba Damo Academy (2018/11-2019/06), Supervised by Lidong Bing and Xiaozhong Liu
- Tencent AI Lab (2018/04-2018/06), Supervised by Yan Song and Lidong Bing
- Georgia Institute of Technology (2014/01-2014/05)
5. Tutorials and Talks
- Juntao Li and Rui Yan. Creative and Artistic Writing via Text Generation. In AAAI’20. Tutorial. [Slide]. (CCF A)
- Juntao Li and Rui Yan. Creative and Artistic Writing via Text Generation. In IJCAI’19. Tutorial. [Slide]. (CCF A)
- YSSNLP2022, Deep Learning for Dialogue System: Chit-Chat and Beyond, [video][Slide]
6. Alumni
- Chenchen Dai (Undergraduate student of Soochow University > Algorithm Engineer, Ant Group(蚂蚁金服,算法工程师))
- Yi Deng (Undergraduate student of Soochow University > University College London (UCL), Master)
7. Selected Awards
- Startrack Visiting Scholars (铸星计划), Microsoft Research Aisa
- Outstanding Graduates Awards (优秀毕业生), Peking University
- AAAI Tutorial Grant
- National Scholarship (国家奖学金), Peking University
8 Academic Services (Program Committee or Reviewer)
- ACL Rolling (Area Chair/Action Editor)
- 2023: ICLR, IJCAI (Senior PC), AACL (Website chair)
- 2022: AAAI, IJCAI, NAACL, KDD, ICLR, EMNLP (Area Chair)
- 2021: AAAI, IJCAI (Senior PC), NAACL-HLT, EACL, ACL (Area Chair), ICONIP (Tutorial, Session Chair)
- 2020: ACL, AAAI, COLING, EMNLP, AACL, CCL
- IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI/PAMI)
- IEEE Transactions on Neural Networks and Learning Systems (TNNLS),
- Computational Linguistics (CL)
- Transactions on Knowledge and Data Engineering (TKDE)