I am an undergraduate researcher in Computer Science at Shanghai Jiao Tong University, in the Zhiyuan Honors Program and MVIG Lab.
I work on:
- embodied intelligence
- multimodal models
- VLA systems
- multimodal evaluation and research tooling
- Building embodied systems with wearable gripper fingertips, force-aware sensing, teleoperation, and VLA-style training.
- Working on multimodal generation, unified understanding, and world-model-adjacent systems.
- Building benchmarks and evaluation pipelines for multimodal models.
- UniG2U: Benchmarking When Generation Helps Understanding in Multimodal Unified Models
- DANet: A RAG-inspired Dual Attention Model for Few-shot Time Series Prediction
- Tri-MARF: A Tri-Modal Multi-Agent Responsive Framework for Comprehensive 3D Object Annotation
- VL-R1-X: Incentivizing Diverse Multimodal Reasoning via Cross-modality Guidance
- StepRouter: From Effort Priors to Utility Posteriors
- GitHub: @nssmd
- Email: 2581235653@sjtu.edu.cn
- Project page: UniG2U