Scholar
Jindong Jiang
Google Scholar ID: 6oo8xOQAAAAJ
Research Scientist, NVIDIA Research
Citations & Impact
Citations
1,169
H-index
9
i10-index
9
Publications
14
Co-authors
10
list available
Contact
Publications
Resume (English only)
Academic Achievements
- May 2025: Preprint 'Token-Efficient Long Video Understanding for Multimodal LLMs' released on arXiv
- Sep 2024: 'Slot State Space Models' accepted to NeurIPS 2024
- Feb 2024: 'Layout-Agnostic Scene Text Image Synthesis with Diffusion Models' accepted to CVPR 2024
- Sep 2023: 'Object-Centric Slot Diffusion' accepted to NeurIPS 2023 as a Spotlight paper (top 3%)
- Apr 2023: Related work accepted to CVPR 2023 GCV Workshop
- 2020: 'Generative Neurosymbolic Machines' accepted to NeurIPS 2020 as Spotlight (top 4%)
- 2020: Co-authored 'Improving Generative Imagination in Object-Centric World Models' published at ICML 2020
- 2020: Co-authored 'SCALOR: Generative World Models with Scalable Object Representations' published at ICLR 2020
Background
- Currently a Research Scientist at NVIDIA Research
- Research interests lie at the intersection of representation learning and visual reasoning
- Focused on developing novel architectures to enhance agents' visual reasoning capabilities
- Long-term goal is to build AI agents capable of human-like reasoning, including uncovering latent structures of the physical world, predicting future scenarios, inferring causality or correlation between events, and performing logical planning
- Current focus areas include Multimodal LLMs, Vision Foundation Models, and their synergy
