About
I am a Ph.D. student in Computer Science at Singapore Management University, supervised by Professor Pradeep Varakantham. My research focuses on training reinforcement learning (RL) agents and humans to be more generalizable, with an emphasis on improving training effectiveness and efficiency. For an example of my recent work, see UTSD.

Before joining SMU, I completed my MSc in Electrical Engineering at the Viterbi School of Engineering, University of Southern California, where I discovered my passion for machine learning and AI research. Outside of my academic pursuits, I enjoy sports, travelling, and supporting FC Bayern Munich.


News
  • [09/2024] My paper has been accepted to NeurIPS 2024 for an oral presentation!

  • [06/2024] I started my internship at Huawei, researching LLM Agents.

  • [08/2020] I started my Ph.D. journey at SMU.


Conference Papers
Improving Environment Novelty Quantification for Effective Unsupervised Environment Design
Jayden Teoh Jing Xiang*, Wenjun Li*, Pradeep Varakantham
Advances in Neural Information Processing Systems (NeurIPS), 2024 (Oral) (Paper) (Website) (Code Releasing Soon )


Unsupervised Training Sequence Design: Efficient and Generalizable Agent Training
Wenjun Li, Pradeep Varakantham
Association for the Advancement of Artificial Intelligence (AAAI), 2024
(Paper) (Website)


Generalization through Diversity: Improving Unsupervised Environment Design
Wenjun Li, Pradeep Varakantham, Dexun Li
International Joint Conference on Artificial Intelligence (IJCAI), 2023
(Paper) (Website)


Facilitating Human-Wildlife Cohabitation through Conflict Prediction
Susobhan Ghosh, Pradeep Varakantham, Aniket Bhatkhande, Tamanna Ahmad, Anish Andheria, Wenjun Li, Aparna Taneja, Divy Thakkar, Milind Tambe
Association for the Advancement of Artificial Intelligence (AAAI), 2022
(Paper)



Preprints
Adaptive Tool Use in Large Language Models with Meta-Cognition Trigger
Wenjun Li, Dexun Li, Kuicai Dong, Cong Zhang, Hao Zhang, Weiwen Liu, Yasheng Wang, Ruiming Tang, Yong Liu
(ArXiv)
Unlocking the Planning Capabilities of Large Language Models with Maximum Diversity Fine-tuning
Wenjun Li, Changyu Chen, Pradeep Varakantham
(ArXiv)
RL as a Surrogate: Teacher Algorithms for Human Learning
Wenjun Li*, Sidney Tio*, Ramesha Karusena, Jimmy Ho, Pradeep Varakantham
(Pending for release)
Marginal Benefit Driven RL Teacher for Unsupervised Environment Design
Dexun Li, Wenjun Li, Pradeep Varakantham
(ArXiv)



Workshop Papers
Generalizable Policy through Diversity: Improving Unsupervised Environment Design
Wenjun Li, Pradeep Varakantham, Dexun Li
Environment Generation for Generalizable Robots, RSS, 2023



Employment
Huawei - Intern, Search and Recommendation Lab - (2024)


Geely - Intern, Algorithm Engineer - (2019)



Education
Singapore Management University - Ph.D. in Compute Science - (2020.08-)


University of Southern California - MSc in Eletrical Engineering - (2018.08-2020.05)


China Jiliang University - BSc in Optoelectronic Information Science and Engineering - (2014.09-2018.06)



Honors
  • Graduate: SMU Presidential Doctoral Fellowship 2023; SMU Presidential Doctoral Fellowship 2024.

  • Undergraduate: Excellent Graduate; 1st Class Scholarship for Academic Excellence.


Invited Talk
  • "Training Robust Reinforcement Learning Agents via Environment Generation" at Telenor Research.


Additional Information
  • Languages: Mandarin, English, German.

  • Reviewing: ICLR 2024, IJCAI 2023.

  • Interests: Football, Travelling, Reading.