cross-encoder/ms-marco-TinyBERT-L-4

The model can be used for Information Retrieval: Given a query, encode the query will all possible passages (e.g. retrieved with ElasticSearch) Then sort the passages in a decreasing order . The training code is available here: SBERT.net Training MS Marco with the training code . Pre-trained Cross-Encoders have been used on the TREC Deep Learning 2019 and the MS Marco Passage Reranking dataset . The performance of these models has been significantly improved on a V100 GPU using Rust tokenizer based on the standard Python tokenizer from Huggingface in version 3.0 . The model was trained on the task using General_TinyBERT_,