Scholar

Yao (Mark) Mu

Google Scholar ID: HK4x3fkAAAAJ

Shanghai Jiao Tong University

Embodied AIReinforcement LearningRoboticsComputer Vision

Citations & Impact

All-time

Citations

2,076

H-index

i10-index

Publications

Co-authors

list available

Contact

Publications

20 items

Browse publications on Google Scholar (top-right) ↗

Resume (English only)

Academic Achievements

Paper 'Skilldiffuser: Interpretable hierarchical planning via skill abstractions in diffusion-based task execution' accepted by CVPR 2024
Three papers (SPET, Tree-Planner, Aligndiff) accepted by ICLR 2024
Paper 'EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought' accepted by NeurIPS 2023 (Spotlight)
Paper 'AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners' accepted by ICML 2023 (Oral Presentation)
Paper 'MetaDiffuser: Diffusion Model as Conditional Planner for Offline Meta-RL' accepted by ICML 2023
Paper 'EC²: Emergent Communication for Embodied Control' accepted by CVPR 2023
Paper 'CO³: Cooperative Unsupervised 3D Representation Learning for Autonomous Driving' accepted by ICLR 2023
Paper 'EUCLID: Towards Efficient Unsupervised Reinforcement Learning with Multi-choice Dynamics Model' accepted by ICLR 2023
Paper 'DOMINO: Decomposed Mutual Information Optimization for Generalized Context in Meta-Reinforcement Learning' accepted by NeurIPS 2022 (Spotlight)
Paper 'CtrlFormer: Learning Transferable State Representation for Visual Control via Transformer' accepted by ICML 2022
Recipient of Hong Kong PhD Fellowship Scheme (HKPFS)
Recipient of HKU Presidential PhD Scholar Programme (HKU-PS)
Student Best Paper Award at the 20th ICCAS
Finalist for Best Student Paper Award at IV2021 (3 out of 450 submissions)

Background

Tenure-track Assistant Professor at AI Institute, Shanghai Jiao Tong University
Research interests: Embodied AI, Generative Robot Agents, Reinforcement Learning, Robotics Control
Research goal: Build a general embodied AI system capable of efficiently learning optimal policies and generalizing well to unseen tasks and environments
Research slogan: Shape the intelligence, Spirit the machine!
Collaborates closely with Prof. Xiaokang Yang (SJTU), Prof. Ping Luo (HKU), and Prof. Mingyu Ding (UNC)

Co-authors

19 total