Unified NLP Models
Bryan discusses his early efforts at Salesforce to create a unified model for natural language processing, focusing on tasks like summarization, translation, and question answering. Despite initial challenges with datasets and architectures, he aimed to establish connections between tasks, emphasizing the potential of transfer learning beyond just word vectors. The conversation highlights the evolution of NLP techniques leading up to the adoption of transformers.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
835: AI Systems as Productivity Engines — with You.com’s Bryan McCann
Related Questions