Model reference · open weights
wiki-all-8-4-multi-dpr2-passage-encoder is an open-weight embedding model from castorini.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the castorini/wiki-all-8-4-multi-dpr2-passage-encoder model card, 2026-10-01
What it is
| Released by | castorini |
|---|---|
| Released | 2023-01-04 |
From the model card
Dense passage retriever (DPR) is a dense retrieval method described in the following paper:
Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, Wen-tau Yih. Dense Passage Retrieval for Open-Domain Question Answering. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6769-6781, 2020.
We have trained our own DPR models with our Wikipedia corpus variants using the Tevatron library.
Our own efforts are described in the paper entitled:
Pre-Processing Matters! Improved Wikipedia Corpora for Open-Domain Question Answering.
This is the passage encoder portion of a 2nd iteration DPR model for the wiki-all-8-4 corpus variant trained on the amalgamation of the NQ, TriviaQA, WQ, and CuratedTREC datasets.
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.