Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the multifaceted world of text embeddings beyond their traditional role in vector search through this 18-minute conference talk. Discover experimental research projects focused on visualizing, manipulating, and interpreting embeddings through a comprehensive journey that begins with fundamental techniques like t-SNE, UMAP, and PCA. Learn how language models can be leveraged to manipulate and interpret embeddings in novel ways. Examine a new tool that enables direct observation of the features that popular embedding models encode into their representations. Gain insights from two years of cutting-edge research that reveals the hidden capabilities and applications of embeddings in AI systems. Understand how embeddings function as more than just search tools and discover their potential for creating AI-augmented interfaces for collaborative work and creativity.
