About Me

I am a researcher at BAAI (Beijing Academy of Artificial Intelligence), a non-profit organization engaged in Al research and development.

My current research interests are focused on Multi-modal LLMs and Embodied Agents.

My supervisors during my Ph.D. study are Prof. Heyan Huang and Assoc. Prof. Yang Gao. I also visited McGill Univerisity as a visiting student for a year advised by Prof. Jackie Chi Kit Cheung.

Download my resumé.

Interests
  • In-context learning
  • Text summarization
  • Low resource language generation
Education

  • Ph.D. Student, 2019 - 2025

    Beijing Institute of Technology

  • B.S., 2015 - 2019

    China University of Geosciences (Beijing)

News

  • [Apr, 2026] ExoActor has been released on ArXiv! [paper]
  • [Apr, 2026] EduBench has been accepted by ACL 2026! [paper]
  • [Feb, 2026] EgoActor has been released on ArXiv! [paper]
  • [Jul, 2025] I have graduated and joined BAAI as a researcher!
  • [Sep, 2024] CItruS has been accepted to be appear at the EMNLP 2024 main conference!
  • [Jun, 2024] A new preprint about the long sequence processing of large language models has been released on ArXiv!
  • [Jan, 2024] A new preprint about the working mechanism of in-context learning has been released on ArXiv!
  • [May, 2023] Finally confirmed to visit the McGill NLP Group and be supervised by Prof. Jackie C. K. Cheung for a year starting on July 1st. Hope to make some academic contributions during my time in Montreal!
  • [Aug, 2022] Another paper on which I am the third author has been accepted by COLING 2022!
  • [Apr, 2022] A paper on which I am the third author has been accepted by IJCAI 2022!
  • [Mar, 2022] A paper has been accepted by SIGIR 2022! See you at the conference!
  • [Jun, 2021] I join Alibaba DAMO academy as a research intern!
  • [Jul, 2021] My volunteer application for ACL 2021 has been accepted. See you at the virtual conference!
  • [May, 2021] A paper has been accepted by ACL-IJCNLP 2021!

Recent Publications

Yanghao Zhou, Jingyu Ma, Yibo Peng, Zhenguo Sun, Yu Bai†, Börje F. Karlsson (2026). ExoActor: Exocentric Video Generation as Generalizable Interactive Humanoid Control. arXiv Preprint.

PDF Cite

Bin Xu*, Yu Bai*, Huashan Sun*, Yiguan Lin*, Siming Liu, Xinyue Liang, Yaolin Li, Yang Gao, Heyan Huang (2026). EduBench: A Comprehensive Benchmarking Dataset for Evaluating Large Language Models in Diverse Educational Scenarios. ACL 2026.

PDF Cite

Yu Bai, MingMing Yu, Chaojie Li, Ziyi Bai, Xinlong Wang, Börje F. Karlsson (2026). EgoActor: Grounding Task Planning into Spatial-aware Egocentric Actions for Humanoid Robots via Visual-Language Models. arXiv Preprint.

PDF Cite

Huashan Sun, Shengyi Liao, Yansen Han, Yu Bai, Yang Gao, Cheng Fu, Weizhou Shen, Fanqi Wan, Ming Yan, Ji Zhang, Fei Huang (2026). SoLoPO: Unlocking Long-Context Capabilities in LLMs via Short-to-Long Preference Optimization. ICLR 2026.

Cite

Yu Bai, Heyan Huang, Cesare Spinoso-Di Piano, Marc-Antoine Rondeau, Sanxing Chen, Yang Gao, Jackie Chi Kit Cheung (2026). Identifying and Analyzing Task-Encoding Tokens in Large Language Models. AAAI 2026.

PDF Cite

Haoqi Yuan, Yu Bai, Yuhui Fu, Bohan Zhou, Yicheng Feng, Xinrun Xu, Yi Zhan, Börje F. Karlsson, Zongqing Lu (2025). Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skills. arXiv Preprint.

PDF Cite

Heyan Huang, Yinghao Li, Huashan Sun, Yu Bai, Yang Gao (2024). How Far Can In-Context Alignment Go? Exploring the State of In-Context Alignment. Findings of EMNLP 2024.

Cite

Jiawei Li, Yizhe Yang, Yu Bai, Xiaofeng Zhou, Yinghao Li, Huashan Sun, Yuhang Liu, Xingpeng Si, Yuhao Ye, Yixiao Wu, Yiguan Lin, Bin Xu, Ren Bowen, Chong Feng, Yang Gao, Heyan Huang (2024). Fundamental Capabilities of Large Language Models and their Applications in Domain Scenarios: A Survey. ACL 2024.

Cite

Yu Bai*, Xiyuan Zou*, Heyan Huang, Sanxing Chen, Marc-Antoine Rondeau, Yang Gao, Jackie Chi Kit Cheung (2024). CItruS: Chunked Instruction-aware State Eviction for Long Sequence Modeling. EMNLP 2024.

PDF Cite Code

Yaolin Li, Yang Gao, Yu Bai, Heyan Huang (2023). Enhancing Consistency with the Fusion of Paralleled Decoders for Text Generation. Information Fusion.

Cite

Jiaao Zhan, Qian Chen, Boxing Chen, Wen Wang, Yu Bai, Yang Gao (2023). DePA: Improving Non-autoregressive Translation with Dependency-Aware Decoder. IWSLT 2023.

PDF Cite

Yu Bai, Heyan Huang, Kai Fan, Yang Gao, Yiming Zhu, Jiaao Zhan, Zewen Chi, Boxing Chen (2022). Unifying Cross-lingual Summarization and Machine Translation with Compression Rate. SIGIR 2022.

PDF Cite Code

Heyan Huang, Yu Bai, Yang Gao, Minpeng Liao (2022). Dynamic Token Halting for Efficient Abstractive Summarization with Importance-aware Regularization. KBS.

Cite

Zewen Chi, Heyan Huang, Luyang Liu, Yu Bai, Xiaoyan Gao, Xian-Ling Mao (2022). Can Pretrained English Language Models Benefit Non-English NLP Systems in Low-Resource Scenarios?. TASLP.

PDF Cite

Xiaochen Liu, Yang Gao, Yu Bai, Jiawei Li, Yinan Hu, Heyan Huang, Boxing Chen (2022). PSP: Pre-trained Soft Prompts for Few-Shot Abstractive Summarization. COLING 2022.

PDF Cite

Jiaao Zhan, Yang Gao, Yu Bai, Qianhui Liu (2022). Stage-wise Stylistic Headline Generation: Style Generation and Summarized Content Insertion. IJCAI-2022.

PDF Cite

Yu Bai, Yang Gao, Heyan Huang (2021). Cross-Lingual Abstractive Summarization with Limited Parallel Resources. ACL-IJCNLP 2021.

PDF Cite Code

Haonan Wang, Yang Gao, Yu Bai, Mirella Lapata, Heyan Huang (2021). Exploring Explainable Selection to Control Abstractive Summarization. AAAI 2021.

PDF Cite Code

Mucheng Ren, Heyan Huang, Ran Wei, Hongyu Liu, Yu Bai, Yang Wang, Yang Gao (2019). Multiple perspective answer reranking for multi-passage reading comprehension. NLPCC 2019.

PDF Cite Code

Contact