Hi, Iām Yanru Wu (å“彦å), a senior undergraduate at Huazhong University of Science and Technology (HUST). I am currently a research intern in the Computational Cognition, Vision, and Learning Lab at Johns Hopkins University, advised by Prof. Alan Yuille and Dr. Jieneng Chen. Previously, I worked with Prof. Yue Wang and Dr. Jiageng Mao in the Physical Superintelligence Lab at the University of Southern California, and with Prof. Yao Wan in ONE Lab at HUST.
My research focuses on building generalizable robot learning systems that connect foundation models with physical feedback and policy adaptation. I am particularly interested in cross-platform robot learning, high-precision and contact-rich manipulation, and learning from visual, force/torque, and tactile signals.
News
- 2026.07Joined the CCVL Lab at Johns Hopkins University as a research intern.
- 2026.03Released our Large Reward Models preprint on visual reward generation for robot policy refinement.
- 2025.05One paper was accepted to Findings of ACL 2025.
- 2025.01One paper was accepted to ICLR 2025 as a Spotlight.
Publications
Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models
Yanru Wu, Weiduo Yuan, Ang Qi, Vitor Guizilini, Jiageng Mao, Yue Wang
A vision-language-model-based framework that generates online visual rewards for robot policy refinement across tasks and environments.
Interleaved Scene Graph for Interleaved Text-and-Image Generation Assessment
Dongping Chen*, Ruoxi Chen*, Shu Pu*, Zhaoyi Liu*, Yanru Wu*, Caixi Chen*, Benlin Liu, Yue Huang, Yao Wan, Pan Zhou, Ranjay Krishna
A benchmark centered on interleaved scene graphs for evaluating models that generate and reason across alternating text and images.
The Impact of Large Language Models in Academia: From Writing to Speaking
Mingmeng Geng*, Caixi Chen*, Yanru Wu*, Yao Wan, Pan Zhou, Dongping Chen
A large-scale analysis of how LLM adoption is changing linguistic patterns in academic papers and conference talks.
Research Experience
Johns Hopkins University Ā· CCVL Lab
University of Southern California Ā· Physical Superintelligence Lab
Huazhong University of Science and Technology Ā· ONE Lab
Education
Huazhong University of Science and Technology
Service
- Reviewer, CVPR Workshop on Synthetic Data for Computer Vision (SynData4CV)