Karan Desai

We introduce nocaps, the first large-scale benchmark for novel object captioning, containing nearly 400 novel object classes, and provide analysis to guide future work. Expand

We propose a new class of probabilistic neural-symbolic models, that have symbolic functional programs as a latent, stochastic variable. Expand

We propose and openly release CRLMaze, a new benchmark for learning continually through reinforcement in a complex 3D non-stationary task based on ViZDoom and subject to several environmental changes. Expand

We propose VirTex -- a pretraining approach using semantically dense captions to learn visual representations. Expand

We propose Contrastive Attention-Supervised Tuning (CAST), a training method to improve the visual grounding ability of contrastive SSL models and propose a solution to overcome them. Expand