work Automated image generation via prompt-based guidance a user-driven T2I framework using prompt engineering with diffusion-based models for complex object generation academic Few-shot Learning for Energy Detection a few-shot learning model to detect energy infrastructure (nuclear, solar, hydro) in aerial imagery. EcoNet Weather Forecasting Ranked 1st in class of 25 groups with 5 different ML models Evaluation of various BERT algorithms evaluated 5 BERT models across 8 GLUE tasks, identifying BERT Uncased as the top performer Pneumonia Detection a DenseNet model to detect pneumonia with bounding boxes, reaching 80% accuracy Terrian Classification a model with CNN and LSTM using time-series data up to 93% accuracy on the test set to improve robotic limb
