MSMARCO Models (Version 3)¶
MS MARCO is a large scale information retrieval corpus that was created based on real user search queries using Bing search engine. The provided models can be used for semantic search, i.e., given keywords / a search phrase / a question, the model will find passages that are relevant for the search query.
The training data constist of over 500k examples, while the complete corpus consist of over 8.8 Million passages.
from sentence_transformers import SentenceTransformer, util model = SentenceTransformer('msmarco-distilroberta-base-v3') query_embedding = model.encode('How big is London') passage_embedding = model.encode('London has 9,787,426 inhabitants at the 2011 census') print("Similarity:", util.pytorch_cos_sim(query_embedding, passage_embedding))
For more details on the usage, see Applications - Information Retrieval
Performance is evaluated on TREC-DL 2019, which is a query-passage retrieval task where multiple queries have been annotated as with their relevance with respect to the given query. Further, we evaluate on the MS Marco Passage Retrieval dataset.
As baseline we show the results for lexical search with BM25 using ElasticSearch.
|Approach||NDCG@10 (TREC DL 19 Reranking)||MRR@10 (MS Marco Dev)|
msmarco-roberta-base-ance-fristp is the MSMARCO Dev Passage Retrieval ANCE(FirstP) 600K model from ANCE. This model should be used with dot-product instead of cosine similarity.
Changes in v3¶
The models from v2 have been used for find for all training queries similar passages. An MS MARCO Cross-Encoder based on the electra-base-model has been then used to classify if these retrieved passages answer the question.
If they received a low score by the cross-encoder, we saved them as hard negatives: They got a high score from the bi-encoder, but a low-score from the (better) cross-encoder.
We then trained the v2 models with these new hard negatives.