Welcome to Yan Wang’s Homepage
My name is WANG Yan (王琰). I am a principal researcher at Tencent AI Platform Department (also known as the Game AI department of Tencent AI Lab). Proir to that, I served as a research scientist at miHoYo and Tencent AI Lab, working on LLM-driven Role Playing and Retrieval-Augmented Generation. I recieved my Ph.D. degree at City University of Hong Kong (CityU) in July 2017, under the instructions of Prof. Hanxiong Li. You can reach me through yanwang.branden(at)gmail.com
News
- 2024/08: A survey paper: “The Oscars of AI Theater: A Survey on Role-Playing with Language Models”, released on Arxiv.
- 2024/07: Our paper “With Greater Text Comes Greater Necessity: Inference-Time Training Helps Long Text Generation” accepted by COLM 2024.
- 2024/04: May be this is the new era of Game Engine: Our paper “Instruction-Driven Game Engines on Large Language Models”, released on Arxiv.
- 2024/01: An interesting paper about long-text generation: “With Greater Text Comes Greater Necessity: Inference-Time Training Helps Long Text Generation”, released on Arxiv.
- 2023/11: Paper “Large Language Models Meet Harry Potter: A Bilingual Dataset for Aligning LLMs with Characters”, accepted by EMNLP findings, see you guys in Singapore!
- 2023/05: Demo paper, “Effidit: An Assistant for Improving Writing Efficiency”, accepted by ACL 2023.
- 2023/01: One milestone pape, “Copy is All You Need”, accepted by ICLR 2023.
- 2022/09: One papers accepted by Neurips 2022.
- Organize a tutorial on Retrieval-Augmented Text Generation at CCL 2022
- 2022/06: Two papers accepted by Interspeech 2022.
- Organized a tutorial on Retrieval-Augmented Text Generation at SIGIR 2022
- Organize a tutorial on Retrieval-Augmented Text Generation at IJCAI 2022
- 2022/04: Our AI writing assitant, Effidit (文涌), was lanuched. Link
- 2021/07: Paper “Neural Machine Translation with Monolingual Translation Memory” was selected as the outstanding paper of ACL 2021.
- 2021/06: Two papers accepted by IEEE Transaction on Games, One paper accepted by TASLP.
- 2021/05: Four long papers accepted by ACL 2021 (3 Main and 1 Findings).
- 2020/09: Two long papers accepted by EMNLP 2020. (Main)
Selected Publications (*=corresponding author)
- Nuo Chen, Yan Wang, Yang Deng, Jia Li. The Oscars of AI Theater: A Survey on Role-Playing with Language Models. Arxiv. 2024. [paper]
- Y. Wang, DM, D. Cai. With Greater Text Comes Greater Necessity: Inference-Time Training Helps Long Text Generation. COLM. 2024. [paper][Social Media]
- Hongqiu Wui, Yan Wang*, Xingyuan Liu, Hai Zhao, Min Zhang. Instruction-Driven Game Engines on Large Language Models. Arxiv. 2024. [paper][Social Media]
Retrieval-Guided Dialogue Generation
- Tian Lan, Deng Cai, Yan Wang*, Heyan Huang, Xian-Ling Mao. Copy is All You Need. ICLR 2023 [paper][Social Media]]
-
Shuming Shi, Enbo Zhao, Duyu Tang, Yan Wang, Piji Li, Wei Bi, Haiyun Jiang, Guoping Huang, Leyang Cui, Xinting Huang, Cong Zhou, Yong Dai, Dongyang Ma. Effidit: Your AI Writing Assistant. ACL 2023 [paper][Product]]
-
Ziqian Dai, Jianwei Yu, Yan Wang*, Nuo Chen, Yanyao Bian, Guangzhi Li, Deng Cai, Dong Yu. Automatic Prosody Annotation with Pre-Trained Text-Speech Model. Interspeech 2022 [paper]
-
Deng Cai, Yan Wang*, Huayang Li, Wai Lam, Lemao Liu. Neural machine translation with monolingual translation memory. Proceedings of ACL 2021. [paper][Social Media]]
-
Yixuan Su, Yan Wang, Simon Baker, Deng Cai, Xiaojiang Liu, Anna Korhonen, Nigel Collier. Prototype-to-Style: Dialogue Generation with Style-Aware Editing on Retrieval Memory. IEEE/ACM Transactions on Audio, Speech, and Language Processing ( Volume: 29). [paper]
-
Deng Cai, Yan Wang*, Wei Bi, Zhaopeng Tu, Xiaojiang Liu, Shuming Shi. Retrieval-guided dialogue response generation via a matching-to-generation framework. Proceedings of EMNLP 2019. [paper]
- Deng Cai, Yan Wang*, Wei Bi, Zhaopeng Tu, Xiaojiang Liu, Shuming Shi. Skeleton-to-response: Dialogue generation guided by retrieval memory. Proceedings of NAACL 2018. [paper]
Multi-turn Dialogue Understanding
-
Yixuan Su, Deng Cai, Qingyu Zhou, Zibo Lin, Simon Baker, Yunbo Cao, Shuming Shi, Nigel Collier, Yan Wang*. Dialogue Response Selection with Hierarchical Curriculum Learning. Proceedings of ACL 2021. [paper
-
Zibo Lin, Deng Cai, Yan Wang*, Xiaojiang Liu, Hai-Tao Zheng, Shuming Shi. The World is Not Binary: Learning to Rank with Grayscale Data for Dialogue Response Selection. Proceedings of EMNLP 2020. [paper]
-
Zhufeng Pan, Kun Bai, Yan Wang*, Lianqiang Zhou, Xiaojiang Liu. Improving Open-Domain Dialogue Systems via Multi-Turn Incomplete Utterance Restoration. Proceedings of EMNLP 2019. [paper]
Persona-based Dialogue Generation
-
Haoyu Song, Yan Wang, Kaiyan Zhang, Wei-Nan Zhang, Ting Liu. BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data. Proceedings of ACL 2021. [paper]
-
Haoyu Song, Yan Wang, Wei-Nan Zhang, Zhengyu Zhao, Ting Liu, Xiaojiang Liu. Profile Consistency Identification for Open-domain Dialogue Agents. Proceedings of EMNLP 2020. [paper]
-
Haoyu Song, Yan Wang, Wei-Nan Zhang, Xiaojiang Liu, Ting Liu. Generate, Delete and Rewrite: A Three-Stage Framework for Improving Persona Consistency of Dialogue Generation. Proceedings of ACL 2020. [paper]
Text Generation
-
Yixuan Su, Tian Lan, Yan Wang, Dani Yogatama, Lingpeng Kong, Nigel Collier. A Contrastive Framework for Neural Text Generation. Neurips 2022. [paper]
-
Yixuan Su, Deng Cai, Yan Wang, David Vandyke, Simon Baker, Piji Li, Nigel Collier. Non-autoregressive Text Generation With Pre-trained Language Models. Proceedings of EACL 2021. [paper]
-
Lei Wang, Yan Wang, Deng Cai, Dongxiang Zhang, Xiaojiang Liu. Translating a Math Word Problem to a Expression Tree. Proceedings of EMNLP 2018. [paper]
-
Yan Wang, Xiaojiang Liu, Shuming Shi. Deep Neural Solver for Math Word Problems. Proceedings of EMNLP 2017. [paper]
Game AI
-
Zelong Yang, Zhufeng Pan, Yan Wang*, Deng Cai, Xiaojiang Liu, Shuming Shi, Shao-Lun Huang. Interpretable real-time win prediction for honor of kings, a popular mobile moba esport. IEEE Transaction on Games.
-
Zelong Yang, Yan Wang*, Piji Li, Shaobin Lin, Shuming Shi, Shao-Lun Huang. Predicting Events in MOBA Games: Dataset, Attribution, and Evaluation. IEEE Transaction on Games.
Speech and NLP
- Ziqian Dai, Jianwei Yu, Yan Wang*, Nuo Chen, Yanyao Bian, Guangzhi Li, Deng Cai, Dong Yu. Automatic Prosody Annotation with Pre-Trained Text-Speech Model. IEEE Interspeech 2022.
Interns
Current Interns
Past Interns
- Hongqiu Wu (PhD student at Shanghai Jiao Tong University)
- Wenyang Hui (Graduate student at ShanghaiTech University)
- Chengyue Jiang (miHoYo, ShanghaiTech University, 05/2023 - 08/2024)
- Xiaotian Qi (miHoYo, Communication University of China, 12/2022 - 08/2024)
- jiaan Wang (miHoYo, Soochow University, 08/2022 - 04/2023)
- Zhiling Zhang (miHoYo, Shanghai Jiao Tong University, 08/2022 - 04/2023)
- Xinting Huang (Tencent AI Lab and Tencent Hunyuan, University of Melbourne, 04/2022 - 08/2022)
- Tian Lan (PhD student at Beijing Institute of Technology, 05/2021 - 06/2023)
- Junxin Li (Tencent and Huawei, Tsinghua University 10/2021-12/2022)
- Dongyang Ma (Tencent AI Lab and miHoYo, Harbin Institute of Technology, 05/2021 - 08/2024)
- Nuo chen (PhD student at HKUST, 11/2021 - 09/2022)
- Ziqian Dai (China Telecom, Beijing University, 10/2021 - 05/2022)
- Deng Cai (Tencent AI Lab, CUHK, ACL 2021 Outstanding Paper Award, 02/2018 - 03/2021)
- Zibo Lin (Tencent Wechat, 08/2019 - 03/2021)
- Zelong Yang (China Aerospace Science and Technology Corporation, 08/2019 - 03/2021)
- Haoyu Song (Huawei Topmind(天才少年), Harbin Institute of Technology, 11/2019 - 03/2021)
- Xiangru Tang (PhD student at Yale University, 06/2020 - 09/2020)
- Changying Hao (Baidu, 04/2020 - 09/2020)
- Yixuan Su (Cohere, University of Cambridge, 10/2018 - 08/2019, 05/2020 - 07/2021)
- Jiaxin Pei (PhD student at University of Michigan, 03/2019 - 08/2019)
- Zhufeng Pan (Google Brain, 10/2018 - 09/2019)
- Ziyang Chen (Tencent AI Lab, 03/2018 - 09/2018)
- Lei Wang (PhD student at Singapore Management University, 11/2017 - 03/2018)
Collaborators
- Lingpeng Kong (assistant professor at the University of Hong Kong, Tencent rhinobird cooperation project)
- Ruihua Song (associate professor at Renmin University of China, Tencent rhinobird cooperation project)
- Liang Pang (assistant professor at Institute of Computing Technology, Chinese Academy of Sciences, Tencent rhinobird cooperation project)
- Weinan Zhang (associate professor at Harbin Institute of Technology, Tencent-HIT joint Lab)
Life & Hobbies
As the captain of the Tencent Technology and Engineering Group basketball team, I have won one championship and two runners-up in Tencent. Before that, I served as the captain of the CityU post-graduate basketball team and won seven championships in Hong Kong.
I also served as the Chair of the Fitness Association of Tencent Technology and Engineering Group.
About ten years ago, I dreamed to be a professional E-sports player (RTS game). Now I am still a top-1000 player in the world in Warcraft III Ladder.