ywzhou

Yongwei Zhou (周勇伟)

I obtained my Ph.D. degree from the Department of Computer Science of HIT in March 2024, under the supervision of Prof. Tiejun Zhao. Before that, I have obtained my Master’s and Bachelor’s degrees in Department of Computer Science and Mathematics of HIT. My research focus on NLP, Pretraining for LLMs， Knowledge Reasoning, QA and Semantic Parsing.

Email: ywzhouphd2018@gmail.com

[Google Scholar] [Semantic Scholar] [Resume]

[Last Updated Jan 2025][Designed by Yongwei Zhou]

News

• 07/2025: A paper on BLI has been accepted by the journal Information Fusion 2025.
• 05/2025; Two papers on pretraining has been accepted by ACL findings 2025
• 02/2025: Our Four-Quadrant Multi-Stage Pretraining paper is now publicly available
• 01/2025: Our Preference Curriculum paper for LLM pretraining is now publicly available
• 04/2024: Pleasure to join Meituan.com as FTE
• 09/2023: One paper is accepted to IEEE/ACM Transactions on Audio, Speech, and Language Processing.
• 02/2022: Our UniRPG (EMNLP 2022) achieves the Top-1 on TAT-QA leaderboard.
• 02/2022: Our OPERA++ (NAACL 2022) achieves the Top-1 on DROP leaderboard.
• 01/2021: Our RoR (EMNLP 2021-Findings) achieves the Top-1 on QuAC leaderboard.

Industry and Research Experience

• MeiTuan, Beijing, Apr 2024- Present
  • Full-time employees
  • Pre-Training for LLMs， Data Strategy

Publications

• [Paper][ACL]FRAMES: Boosting LLMs with A Four-Quadrant Multi-Stage Pretraining Strategy
  Xuemiao Zhang^*, Liangyu Xu^*, Feiyu Duan^*, Yongwei Zhou, Sirui Wang, Jingang Wang, Xunliang Cai. ACL 2025 findings

• [Paper] [ACL] Preference Curriculum: LLMs Should Always Be Pretrained on Their Preferred Data.
  Yongwei Zhou^*, Xuemiao Zhang^*, Liangyu Xu^*, Feiyu Duan^*, Sirui Wang, Jingang Wang, Xunliang Cai. ACL 2025 findings

• [Paper][中科院1区]Cross-Lingual Semantic Information Fusion for Word Translation Enhancement.
  Qiuyu Ding, Hailong Cao, Zhiqiang Cao, Yongwei Zhou, Tiejun Zhao. (Information Fusion 2025) (JCR Q1)

• [Paper] [Arxiv] Dual Instruction Tuning with Large Language Models for Mathematical Reasoning.
  Yongwei Zhou, Tiejun Zhao. (arXiv:2403.18295)

• [Paper] [Arxiv] HopPG: Self-Iterative Program Generation for Multi-Hop Question Answering over Heterogeneous Knowledge.
  Yingyao Wang^*, Yongwei Zhou^*, Chaoqun Duan, Junwei Bao, Tiejun Zhao. (arXiv:2308.11257)

• [Paper] [中科院1区| CCF B | 清华A] Operation-Augmented Numerical Reasoning for Question Answering.
  Yongwei Zhou, Junwei Bao, Youzheng Wu, Xiaodong He, Tiejun Zhao.
  IEEE/ACM Transactions on Audio, Speech, and Language Processing(TASLP 2024) (JCR Q1， IF=5.4)

• [Paper] [Code] [CCF-B | 清华A] UniRPG: Unified Discrete Reasoning over Table and Text as Program Generation.
  Yongwei Zhou, Junwei Bao, Chaoqun Duan, Youzheng Wu, Xiaodong He, Tiejun Zhao. EMNLP 2022

• [Paper] [CCF-B | 清华B] OPERA: Operation-Pivoted Discrete Reasoning over Text.
  Yongwei Zhou, Junwei Bao, Chaoqun Duan, Haipeng Sun, Jiahui Liang, Yifan Wang, Jing Zhao, Youzheng Wu, Xiaodong He, Tiejun Zhao. NAACL 2022

• [Paper] [CCF-C] EviDR: Evidence-Emphasized Discrete Reasoning for Reasoning Machine Reading Comprehension.
  Yongwei Zhou, Junwei Bao, Haipeng Sun, Jiahui Liang, Youzheng Wu, Xiaodong He, Bowen Zhou, Tiejun Zhao. NLPCC 2021

• [Paper] [Code] [CCF-B | 清华A] RoR: Read-over-Read for Long Document Machine Reading Comprehension.
  Jing Zhao, Junwei Bao, Yifan Wang, Yongwei Zhou, Youzheng Wu, Xiaodong He, Bowen Zhou. EMNLP (Findings) 2021

Education

• [201809 - 202403] Ph.D., Computer Science, Harbin Institute of Technology
  • Research Interests: NLP, Knowledge Reasoning, QA
  • Advisor: Prof. Tiejun Zhao
  • Lab: Machine Translation and Intelligence (MI&T)
• [2016 - 2018] M.S., Computer Science, Harbin Institute of Technology
  • Research Interests: String Approximate Query
  • Advisor: Prof. Hong Gao
• [2012 - 2016] B.S., Mathematics, Harbin Institute of Technology

Internship Experience

• JD AI Research, Beijing, June 2020 - May 2023
  • Research Intern, QA and Knowledge Reasoning
  • Advisor: Dr.Junwei Bao and Dr. Youzheng Wu
• Noah’s Ark Lab, Huawei, Shenzhen, Oct 2019 - May 2020
  • Research Intern, QA and Knowledge Reasoning
  • Advisors: Dr.Lifeng Shang

Honors & Awards

• [2022] UniRPG achieves the Top-1 on the TAT-QA leaderboard. (JD AI Research)
• [2022] OPERA++ achieves the Top-1 on the DROP leaderboard. (JD AI Research)
• [2021] RoR achieves the Top-1 on the QuAC leaderboard. (JD AI Research)
• [2016] Outstanding Graduates Award (Harbin Institute of Technology)