Yekun Chai

Baidu NLP


chaiyekun (at)

I am a staff research engineer working on large language models (LLMs) at Baidu NLP. Before that, I was affiliated with Institute of Automation, Chinese Academy of Sciences (CASIA). I graduated from Edinburgh Informatics in 2018 under the supervision of Dr. Adam Lopez and Dr. Naomi Saphra.

My research endeavors revolve around the generative pre-training paradigm of NLP, with a particular emphasis on:

  • General LLM pre-training, prompting, instruction tuning, and their variants across tasks, languages, and modalities;
  • Augmented LLMs with non-parametric priors;
  • LLM alignment with human preferences.


Feb 20, 2024 One paper on multilingual code generation benchmarks has been accepted to LREC-COLING 2024.
Jan 16, 2024 One paper on reward modeling has been accepted to ICLR 2024 (Spotlight:sparkles:). Code is available here.
Sep 23, 2023 One paper on XAI has been accepted to NeurIPS 2023 Datasets and Benchmarks Track. Code is available here.
May 02, 2023 Our work on multilingual text and code pre-training (ERNIE-Code) has been accepted to ACL 2023 Findings. Code is available here.

selected publications

    HumanEval-XL: Pioneering Cross-Lingual Code Generation and Multilingual NLP Benchmarking
    Qiwei Peng*Yekun Chai* ,  and  Xuhong Li
    In The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation , May 2024
  2. ICLRSpotlight
    Tool-Augmented Reward Modeling
    Lei Li*Yekun Chai*Shuohuan Wang ,  Yu Sun , and 3 more authors
    In The Twelfth International Conference on Learning Representations (ICLR) , May 2024
  3. ACLFindings
    ERNIE-Code: Beyond English-Centric Cross-lingual Pretraining for Programming Languages
    Yekun ChaiShuohuan Wang ,  Chao Pang ,  Yu Sun , and 2 more authors
    In Findings of the Association for Computational Linguistics: ACL 2023 , Jul 2023
  4. EMNLPFindings
    Clip-Tuning: Towards Derivative-free Prompt Learning with a Mixture of Rewards
    Yekun ChaiShuohuan Wang ,  Yu Sun ,  Hao Tian , and 2 more authors
    In Findings of the Association for Computational Linguistics: EMNLP 2022 , Dec 2022