Zhenyu ZHANG /


Staff Algorithm Engineer at Baidu NLP (Wenxin Team)


 Baidu Technology Park, No.10 Yard, Xibeiwang East Road, Haidian District, Beijing.

 zzysay[at]gmail/163.com

Google Scholar / DBLP

Biography

I am currently a Staff Algorithm Engineer in the Wenxin (文心) Team at Baidu NLP. Before that, I received my Ph.D. degree from University of Chinese Academy of Sciences (UCAS) in 2022, under the supervision of Prof. Tingwen Liu and Prof. Li Guo.

My research interests broadly include Natural Language Processing and Multimodal Models. Now, I am also committed to the frontier exploration and practical application of Large Language Model (LLM) and Artificial Intelligence Generated Content (AIGC), with core contributions to ERNIE series of models like ERNIE 4.0/3.5/3.0 (LLMs), ERNIE-ViLG 2.0 (Text2Image), ERNIE-Layout (Document Intelligent).

I have published several papers in AAAI, ACL, CVPR, KDD, MM, NeurIPS, WWW, etc., and won the Best Paper Award of PAKDD 2020.

News

  • [2025/05] Four papers about Efficient Large Language Model and Parameter-Efficient Fine-Tuning were accepted to ACL 2025.
  • [2025/05] One paper about Efficient Large Language Model was accepted to ICML 2025.
  • [2025/02] One paper about Multimodal Large Language Model was accepted to CVPR 2025.
  • [2024/11] One paper about Large Language Model Evaluation was accepted to COLING 2025.
  • [2024/11] One paper about Text-Video Retrieval was accepted to KDD 2025.
  • [2024/09] One paper about Efficient Large Language Model was accepted to NeurIPS 2024.
  • [2024/05] Two papers about Efficient and Long-Context Language Model were accepted to ACL 2024.
  • [2023/03] 文心一言 (a.k.a. ERNIE Bot), Baidu's Knowledge-Enhanced Large Language Model, was released.
  • [2023/02] One paper (ERNIE-ViLG 2.0) about Text-to-Image Generation was accepted to CVPR 2023.
  • [2023/01] One paper about Information Extraction was accepted to WWW 2023.
  • [2022/10] Two papers about Information Extraction and Document Intelligence (ERNIE-Layout) were accepted to EMNLP 2022.
  • [2022/07] One paper about Document Intelligence was accepted to MM 2022.
  • [2022/02] One paper about Language Model was accepted to ACL 2022.
  • [2021/08] One paper about Named Entity Recognition was accepted to EMNLP 2021.
  • [2021/06] One paper about Relation Extraction was accepted to ECML/PKDD 2021.
  • [2021/05] One paper about Relation Extraction was accepted to ACL 2021.
  • [2021/01] One paper about Information Extraction was accepted to WWW 2021.
  • [2020/09] Two papers about Relation Extraction were accepted to COLING 2020.
  • [2020/09] Two papers about Named Entity Recognition and Event Extraction were accepted to EMNLP 2020.
  • [2020/05] One paper about Relation Extraction was won the Best Paper Award of PAKDD 2020.
  • [2020/01] One paper about Entity Linking was accepted to WWW 2020.
  • [2019/11] One paper about Relation Extraction was accepted to AAAI 2020.
  • [2019/05] One paper about Relation Extraction was accepted to IJCAI 2019.
  • [2019/01] One paper about Entity Linking was accepted to WWW 2019.

Selected Publications ->

Full Publication List

HFT: Half Fine-Tuning for Large Language Models
Tingfeng Hui¡, Zhenyu Zhang†, Shuohuan Wang, Yu Sun, Weiran Xu, Hua Wu
The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Upcycling Instruction Tuning from Dense to Mixture-of-Experts via Parameter Merging
Tingfeng Hui¡, Zhenyu Zhang†, Shuohuan Wang, Yu Sun, Hua Wu, Sen Su
The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

BeamLoRA: Beam-Constraint Low-Rank Adaptation
Naibin Gu*¡, Zhenyu Zhang*, Xiyu Liu, Peng Fu, Zheng Lin, Shuohuan Wang, Yu Sun, Hua Wu, Weiping Wang, Haifeng Wang
The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

ERNIE-ViLG 2.0: Improving Text-to-Image Diffusion Model with Knowledge-Enhanced Mixture-of-Denoising-Experts
Zhida Feng*¡, Zhenyu Zhang*, Xintong Yu*, Yewei Fang, Lanxin Li, Xuyi Chen, Yuxiang Lu, Jiaxiang Liu, Weichong Yin, Shikun Feng, Yu Sun, Li Chen, Hao Tian, Hua Wu, Haifeng Wang
The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2023 (CVPR 2023)

Learning Structural Co-occurrences for Structured Web Data Extraction in Low-Resource Settings
Zhenyu Zhang, Bowen Yu, Tingwen Liu, Tianyun Liu, Yubin Wang, Li Guo
The Web Conference 2023 (WWW 2023)

Layout-Aware Information Extraction for Document-Grounded Dialogue: Dataset, Method and Demonstration
Zhenyu Zhang*, Bowen Yu*, Haiyang Yu, Tingwen Liu, Cheng Fu, Jingyang Li, Chengguang Tang, Jian Sun, Yongbin Li
The 30th ACM International Conference on Multimedia (MM 2022)

From What to Why: Improving Relation Extraction with Rationale Graph
Zhenyu Zhang, Bowen Yu, Xiaobo Shu, Mengge Xue, Tingwen Liu, Li Guo
The 59th Annual Meeting of the Association for Computational Linguistics (ACL 2021, Findings)

Distilling Knowledge from Well-informed Soft Labels for Neural Relation Extraction
Zhenyu Zhang, Xiaobo Shu, Bowen Yu, Tingwen Liu, Jiapeng Zhao, Quangang Li, Li Guo
The 34th AAAI Conference on Artificial Intelligence (AAAI 2020)

Academic Services

Selected Awards

  • 2023. Baidu Pride. Baidu Inc..
  • 2022. AIDU Talent Program. Baidu Inc..
  • 2022. Outstanding Graduates. Beijing Municipal Education Commission.
  • 2021. Best Paper Award Nominee. IEEE DSC.
  • 2021. Pacemaker to Merit Student. University of Chinese Academy of Sciences (UCAS).
  • 2020. National Scholarship for Doctoral Students. Ministry of Education, China.
  • 2020. Best Paper Award. PAKDD.
  • 2017. Outstanding Graduate. Hu'nan Provincial Education Commission.
  • 2017. First Prize of the 3rd National College Competition on Cloud Computing. Ministry of Education, China.
  • 2016. Cyber Security Scholarship. Internet Development Foundation, China.
  • 2016. National Scholarship for Undergraduate Students. Ministry of Education, China.
  • 2015, 2016. Second Prize of the 8th/9th National College Student Information Security Contest. Ministry of Education, China.
  • 2015. Second Prize of the 2nd National College Competition on Internet of Things. Ministry of Education, China.

Work Experiences