Mimic Intent, Not Just Trajectories
Robotics: Science and Systems RSS 2026 2026
We propose learning to mimic the underlying intent of demonstrations rather than directly copying trajectories, enabling more robust and generalizable robot behavior.
Publications
* denotes equal contribution. See also my Google Scholar profile.
Conference Papers
AttentionAR: AR Adaptation and Warning for Real-World Safety via Attention Modeling and MLLM Reasoning
Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology UIST 2025 2025
An augmented reality system that combines attention modeling with multimodal large language model reasoning to improve real-world safety awareness.
Goal-Reaching Policy Learning from Non-Expert Observations via Effective Subgoal Guidance
Conference on Robot Learning CoRL 2024 2024
We propose a subgoal-guided imitation learning framework that enables goal-reaching policy learning from non-expert, suboptimal observations without requiring expert demonstrations.
Diffusion Models as Optimizers for Efficient Planning in Offline RL
European Conference on Computer Vision ECCV 2024 2024
We leverage diffusion models as trajectory optimizers for offline reinforcement learning, achieving efficient planning by treating diffusion sampling as an optimization process.
Multimodal Apology: Using WebXR to Repair Trust with Virtual Companion
2023 IEEE Conference on Virtual Reality and 3D User Interfaces IEEE VR 2023 2023
A multimodal apology system using WebXR to investigate how virtual companions can repair trust after service failures through synchronized verbal and nonverbal cues.
Preprints
Any House Any Task: Scalable Long-Horizon Planning for Abstract Human Tasks
arXiv preprint arXiv 2026
A scalable framework for long-horizon task planning that generalizes across diverse house layouts and abstract human task specifications.