Baptiste L.SPARSIFY IT : Block Sparse Lottery Tickets Hypothesis Pruning — NNs Sparsification Part 1Sparsity got a lot of Attention lately, through Sparse MoEs & the release of Mixtral, and through Pruning technics. While the motivation…Mar 21Mar 21
Baptiste L.Training a Contextual Compression ModelFor RAG, a Contextual Compression step can ameliorate the results in multiple ways :Aug 28, 20231Aug 28, 20231
Baptiste L.Training a Retriever on Synthetic DataToday we are going to use the previously generated synthetic queries to train a retriever on specific type of documents.Aug 24, 2023Aug 24, 2023
Baptiste L.Synthetic Data and BM25 Search SystemIn this article we will see how to make a Search System with a Query Generation model and BM25 and we will evaluate this system.Aug 21, 2023Aug 21, 2023
Baptiste L.Synthetic Queries to Doc Dataset Generation — A News Semantic Search EngineThis is the first article of a Series that will cover advanced Semantic Search Methods & Algorithms. In order to create a Newspaper…Aug 11, 2023Aug 11, 2023
Baptiste L.LangChain Arxiv Tutor : Long Text Summarization, RetrievalQA and Vector DatabasesIn this second Article we will talk about Long Text Summarization, Semantic Search and populate a Vector Database.Jul 7, 20231Jul 7, 20231
Baptiste L.LangChain Arxiv Tutor : Data LoadingThis Series of Articles covers the usage of LangChain, to create an Arxiv Tutor. That will allow anyone to interact in different ways with…Jul 6, 20233Jul 6, 20233