Self-evolving AI agent for climate-science workflows.
Agent workflow design and climate-science reasoning evaluation.
- AI4Earth
- Agent workflow
Self-evolving AI agent for climate-science workflows.
Agent workflow design and climate-science reasoning evaluation.
Large-scale, multi-view 3D human pose dataset and benchmark under real-world conditions.
Lead author; dataset construction, benchmark design, and evaluation.
Instruction-tuning dataset, framework, and benchmark for multimodal large language models.
Core contributor for multimodal instruction data, benchmark design, and framework development.
Transformer-based weakly supervised object localization with implicit spatial calibration.
Co-author; vision representation and localization research.
Reinforcement learning for flexible exposure bracketing selection in HDR imaging.
Co-author; low-level vision and HDR imaging research.