SPLADE v3 DistilBERT SPLADE v3 DistilBERT is the DistilBERT version of naver/splade v3 . For more details, see our arXiv companion book: https://arxiv.org/abs/2403.06789 To use SPLADE, please visit our GitHub repository: https://github.com/naver/splade Performance MRR@10 (MS MARCO dev) avg nDCG@10 (BEIR 13) naver/splade v3 distilbert 38.7 50.0 Model Details This is a SPLADE Sparse Encoder model. It maps sentences & paragraphs to a 30522 dimensional sparse vector space and can be used for semantic search and sparse retrieval. Model Description Model Type: SPLADE Sparse Encoder Base model: distilbert/distilbert base uncased Maximum Sequence Length: 512 tokens (256 for evaluation reproduction) Output Dimensionality: 30522 dimensions Similarity Function: Dot Product Full Model Architecture Usage Direct Usage (Sentence Transformers) First install the Sentence Transformers library: Then you can load this model and run inference. Citation If you use our checkpoint, please cite our work:
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy