I work on grounding language within the visual world with neural models that generate meaningful & useful concept representations from images and text. I’m interested in optimizing such models via methods such as reinforcement learning. Here are my publications. Future-wise, I look to build systems that holistically understand the world and lead to transformative AI.