Luke Merrick

Dec 3, 2024 | arxiv.org | Luke Merrick |Gaurav Nuti |Daniel Campos

arXiv:2412.04506 (cs) View PDF HTML (experimental) Comments: Subjects: Computation and Language (cs.CL); Information Retrieval (cs.IR); Machine Learning (cs.LG) Cite as: arXiv:2412.04506 [cs.CL] (or arXiv:2412.04506v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2412.04506 Submission history From: Luke Merrick [ view email] [v1] Tue, 3 Dec 2024 22:59:36 UTC (528 KB) Bibliographic Tools Bibliographic Explorer Toggle Bibliographic Explorer () Connected Papers Toggle Connected Papers...

Jul 18, 2024 | snowflake.com | Luke Merrick

Today Snowflake released the world’s most pragmatic text embedding model for English-language search: arctic-embed-m-v1.5.  Our new model can deliver tiny embedding vectors — packing nearly 8 million per GB of storage — enabling up to a 24x improvement in retrieval system scalability while still delivering uncompromising retrieval quality. Building on our v1.0 open source release, this model continues to run just as fast as its predecessor and is released with a permissive Apache 2.0 license.

Mar 28, 2024 | medium.com | Luke Merrick

A deep dive into one of the core technologies behind Retrieval Augmented Generation (RAG)Coauthors: Daniel Campos, Danmei XuConsider these two queries: “What is Tom Cruise’s height,” and“How tall is Maverick from Top Gun?” They mean the same thing but lack shared words. How could a computer recognize these as similar?