Hi! I’m Hui Wu, a Senior Engineer at Baidu Inc. I received my Ph.D. from the Aerospace Information Research Institute, Chinese Academy of Sciences, in 2026. Prior to that, I earned my B.E. from the University of Science and Technology Beijing in 2021.

My current research focuses on LLM mid-training, with an emphasis on improving the efficiency and effectiveness of large language model training. I am interested in how training data and training strategies influence model capabilities. Previously, I worked on retrieval-augmented generation (RAG), embodied AI (e.g., Minecraft agents), and downstream NLP tasks.

🔖 Collaboration. Open to research collaborations on related research topics. Welcome to reach out!

🔥 News

• 2026.08:  🎉🎉 My first-author paper is accepted to EMNLP 2026. See you in Budapest, Hungary!
• 2026.05:  🎉🎉 Successfully passed my PhD defense! 🎓
• 2026.04:  🎉🎉 My first-author paper is accepted to ACL 2026. See you in San Diego, USA!
• 2025.11:  🎉🎉 One paper is accepted to AAAI 2026.
• 2025.07:  🎉🎉 My first-author paper is accepted to ACMMM 2025. See you in Dublin, Ireland!
• 2025.04:  🎉🎉 One paper is accepted to TGRS.
• 2025.01:  🎉🎉 My first-author paper is accepted to KBS.

📝 Publications

• MoEC: A Memory-Routed Mixture-of-Experts Controller for Adaptive Minecraft Control
  Hui Wu, Chao Xu, Jianghui Wang, Ziqiong Liu, Dong Li, Yiwei Dai, Emad Barsoum
  ACL 2026 | paper

• Not All Preferences Deserve Gradients: Understanding Gradient Utility in Offline Reasoning Alignment
  Hui Wu, Hengyi Cai, Jinman Zhao, Xinran Chen, Ziheng Li, Zhejun Zhao, Shuaiqiang Wang, Yuchen Li, Dawei Yin
  EMNLP 2026 | coming soon

• MARA: A Multimodal Adaptive Retrieval-Augmented Framework for Document Question Answering
  Hui Wu, Haoquan Zhai, Yuchen Li, Hengyi Cai, Peirong Zhang, Yidan Zhang, Lei Wang, Yingyan Hou, Shuaiqiang Wang, Dawei Yin
  ACM MM 2025 | paper

• CoT-driven Framework for Short Text Classification: Enhancing and Transferring Capabilities from Large to Smaller Model
  Hui Wu, Yuanben Zhang, Zhonghe Han, Yingyan Hou, Lei Wang, Siye Liu, Qihang Gong, Yunping Ge
  Knowledge-Based Systems | paper

• Beyond Step Pruning: Information Theory Based Step-level Optimization for Self-Refining Large Language Models
  Jinman Zhao, Erxue Min, Hui Wu, Ziheng Li, Zexu Sun, Hengyi Cai, Shuaiqiang Wang, Xu Chen, Gerald Penn
  AAAI 2026 | paper

• Language-guided Object Localization via Refined Spotting Enhancement in Remote Sensing Imagery
  Peirong Zhang, Yidan Zhang, Hui Wu, Xiaoxuan Liu, Yingyan Hou, Lei Wang
  IEEE Transactions on Geoscience and Remote Sensing | paper

🎖 Honors and Awards

• 2026, Beijing Municipal Outstanding Graduate
• 2025, Merit Student,Chinese Academyof Sciences

📖 Educations

• 2021.09 – 2026.06, Ph.D., Aerospace Information Research Institute, Chinese Academy of Sciences
• 2017.09 – 2021.06, B.S., University of Science and Technology Beijing

📚 Academic Services

• Conference Reviewer: Neurips, ACL, EMNLP, ACMMM
• Journal Reviewer: TGRS, KBS