Sonam Pankaj

ColPali is revolutionary—here’s why: it combines document retrieval with a vision-based large language model, allowing you to search directly within images without needing to extract text.

Vector databases are everywhere, powering LLMs. But indexing embeddings, especially multivector embeddings like ColPali and Colbert, at a bulk is memory intensive.

Language models are becoming bigger and bigger breaking benchmarks like Blue and Blurb scores, but accuracy should not be the only thing that should be considered, but societal biases and inclusion as well. In this talk, we will look deeply into if these models are biased how we can mitigate...