World Event
Research on "World Events" currently focuses on leveraging large datasets and advanced machine learning models to understand and predict various global phenomena. This includes using transformer-based architectures and graph neural networks to analyze multimodal data (images, text, sensor readings) for tasks such as predicting wildfire risk, optimizing traffic flow, and forecasting e-commerce demand. These efforts aim to improve the accuracy and robustness of predictions, particularly in handling anomalies and diverse geographical contexts, leading to more effective resource allocation and decision-making across various sectors. The ultimate goal is to develop more comprehensive and reliable models for understanding complex global systems and their interactions.
Papers
VIA: Unified Spatiotemporal Video Adaptation Framework for Global and Local Video Editing
Jing Gu, Yuwei Fang, Ivan Skorokhodov, Peter Wonka, Xinya Du, Sergey Tulyakov, Xin Eric Wang
Towards Understanding Domain Adapted Sentence Embeddings for Document Retrieval
Sujoy Roychowdhury, Sumit Soman, H. G. Ranjani, Vansh Chhabra, Neeraj Gunda, Shashank Gautam, Subhadip Bandyopadhyay, Sai Krishna Bala