Yun Li received his Ph.D. from the Graduate School of Information Science and Technology, The University of Tokyo in September 2026, under the supervision of Prof. Manabu Tsukada. His research focuses on safe and efficient decision-making for autonomous driving, including diffusion-based trajectory planning and safety alignment of vision-language-action (VLA) models via preference learning, with closed-loop validation and real-vehicle deployment on Autoware/AWSIM. During his Ph.D., he was a JSPS Research Fellow (DC2) and a Student Researcher at TIER IV.

Publications (12)
Int'l conferenceCARE: Camera-Residual Reserves for First Sightings in Adaptive LiDAR Sensing
Jiachen Gong, Yun Li, Ehsan Javanmardi, Wencan Mao, Manabu Tsukada — The 18th Asian Conference on Computer Vision (ACCV 2026), Osaka, Japan, 2026
Int'l conferenceTo Route or to Adapt? Multi-Expert Diffusion Planners in Unseen Regions
PosterNL2TL-Tree: Natural Language to Nested Temporal-Logic Trees for Robustness-Guided Flow-Matching Planning
Jiachen Gong, Wencan Mao, Ehsan Javanmardi, Yun Li, Quanxi Zhou, Manabu Tsukada — Companion of the 2026 ACM International Joint Conference on Pervasive and Ubiquitous Computing and the 2026 ACM International Symposium on Wearable Computers (UbiComp/ISWC '26), Poster, Shanghai, China, 2026
ThesisPreference-Aligned Generative Models for Safe and Efficient Autonomous Driving: Vision-Language-Action Models and Real-Time Diffusion Planning
Yun Li — Ph.D Thesis, The University of Tokyo, 2026
JournalCausal Scene Narration: Inference-Time Text Restructuring for Vision-Language-Action Driving
Yun Li, Yidu Zhang, Simon Thompson, Ehsan Javanmardi, Manabu Tsukada — Transactions on Machine Learning Research, 2026
Int'l conferenceAn Open-Source Modular Benchmark for Diffusion-Based Motion Planning in Closed-Loop Autonomous Driving
Yun Li, Simon Thompson, Yidu Zhang, Ehsan Javanmardi, Manabu Tsukada — The IEEE International Conference on Intelligent Transportation Systems (ITSC2026), Naples, Italy, 2026
Workshop / WiPLearning Diffusion Planners from World Feedback: A No-Go Result on Bit-Exact Safety Rewards and an ODD-Adaptive Shared/Expert Decomposition
Domestic正則化付き多重選好学習による自動運転VLAモデルの安全制約アライメント
Int'l conferenceMulti-PrefDrive: Optimizing Large Language Models for Autonomous Driving Through Multi-Preference Tuning
Int'l conferencePrefDrive: Enhancing Autonomous Driving through Preference-Guided Large Language Models
Int'l conferenceLarge Language Models for Human-like Autonomous Driving Decision Making: A Survey
