Learning Visiolinguistic Representations with ViLBERT w/ Stefan Lee - #358
Published:Mar 18, 2020 21:04
•1 min read
•Practical AI
Analysis
This article summarizes a podcast episode of Practical AI featuring Stefan Lee, an assistant professor at Oregon State University. The episode focuses on Lee's research paper, ViLBERT, which explores pretraining task-agnostic visiolinguistic representations for vision-and-language tasks. The discussion likely covers the model's development, training process, and the adaptation of BERT models to incorporate visual information. The conversation also touches upon the future of integrating visual and language tasks, indicating a focus on the intersection of computer vision and natural language processing. The episode provides insights into the creation and application of a model designed to bridge the gap between visual and textual data.
Key Takeaways
- •The episode features Stefan Lee discussing his research on ViLBERT.
- •ViLBERT focuses on pretraining visiolinguistic representations for vision-and-language tasks.
- •The discussion covers model development, training, and the integration of visual information with language models.
Reference
“We discuss the development and training process for this model, the adaptation of the training process to incorporate additional visual information to BERT models, where this research leads from the perspective of integration between visual and language tasks.”