Research

🔬 Research

Generative AI, video-language understanding, efficient deep learning, and machine learning for sustainability.

Hiring at AMD: We are hiring for full-time positions at all levels working on generative AI applications on AMD hardware.

🎥 Video-Language & Long-Form Video

Foundational models and systems for long-form video understanding, captioning, ranking, and mitigating hallucinations in video-language models.

âš¡ Efficient & Multimodal Models

Dynamic inference, structured pruning, weight sharing, and lightweight detectors for grounding-based vision-and-language models.

🌱 Computational Sustainability

Remote sensing, geolocated data, self-supervised learning, and interpretable models for agriculture, poverty mapping, and environmental monitoring.

🎨 Generative Models

Generative modeling for satellite imagery, data augmentation, and counterfactual video generation for robust VLM evaluation.

See the full publication list or Google Scholar profile for a complete bibliography.