Scholar
Yijun Yang
Google Scholar ID: X0quXnsAAAAJ
Tencent AI Lab
Citations & Impact
Citations
219
H-index
8
i10-index
7
Publications
20
Co-authors
6
list available
Publications
Resume (English only)
Academic Achievements
- 1. September 2025, one paper (World model-based LLM Agent, WALL-E) was accepted by NeurIPS 2025.
- 2. June 2025, one paper (VLM Agents Trained with RL) was accepted by ICCV 2025.
- 3. May 2025, one paper (findings) was accepted by ACL 2025.
- 4. March 2025, identified a critical challenge in RL-based VLM agent training and proposed an innovative solution Guided Thought Reinforcement (GTR) that combines the strengths of RL and IL.
- 5. February 2025, awarded the 2024 DEAN’S LIST by UTS Faculty of Engineering & Information Technology for excellent PhD thesis and research achievements.
- 6. January 2025, released a new benchmark MM-IQ for assessing the core reasoning capabilities of large multimodal models.
- 7. November 2024, one paper (A Python Library for Black-box Optimization) was accepted by JMLR MLOOS.
Research Experience
- Works as a researcher at Tencent AL Lab, focusing on training generalist AI agents in complex environments through reinforcement and imitation learning, leveraging the rich prior knowledge from pretrained foundation models such as LLMs, VLMs, and video generation models. Gained valuable industry experience as a research intern at JD Explore Academy, under the mentorship of Prof. Li Shen from Sun Yat-sen University, and within the team led by Dr. Xiaodong He. Also contributed to the development of advanced multimodal models as a researcher at Tencent Hunyuan, working under the guidance of Dr. Wei Liu.
Education
- Ph.D. in Computer Science; University: University of Technology Sydney (UTS); Advisor: Prof. Chengqi Zhang; Time: Not specifically mentioned; Major: Computer Science.
Background
- Research Interests: Reinforcement Learning, Machine Learning, and LLM/VLM-based AI Agents. Brief Introduction: A researcher at Tencent AL Lab, has published papers at top-tier AI/ML venues such as ICML, ICLR, NeurIPS, CVPR, ICCV, etc.
Miscellany
- Personal interests and other information not explicitly provided.
