Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers
Source
huggingface.co
Date
Terms in this piece · Glossary
fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
Why it matters
Late-interaction retrievers are now trainable in Sentence Transformers, and a single-GPU day of fine-tuningTaking a trained model and training it a bit more on your own examples so it gets better at one specific job.Full definition → beat general-purpose dense, sparse and lexical retrievers on domain data.
articleDenseOn with the LateOn: Fully Open Dense and Late-Interaction Models for Multilingual, Long-Context, and Code SearchRapha\"el Sourty, Antoine Chaffin, Paulo Roberto Moura Junior, Am\'elie Chatelain
articleAttention Is All You NeedAshish Vaswani et al.
articleSwitch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient SparsityWilliam Fedus et al.