MrLight
/

dse-phi35-vidore-ft

Model card Files Files and versions Community

MrLight commited on Sep 6

Commit

8c02be2

•

1 Parent(s): b9550c5

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -20,7 +20,7 @@ DSE-Phi3-Vidore-ft is a bi-encoder model designed to encode document screenshots
 The model, `Tevatron/dse-phi35-vidore-ft`, is trained using 1/10 of the `Tevatron/docmatix-ir` dataset, a variant of `HuggingFaceM4/Docmatix` specifically adapted for training PDF retrievers with Vision Language Models in open-domain question answering scenarios. For more information on dataset filtering and hard negative mining, refer to the [docmatix-ir](https://huggingface.co/datasets/Tevatron/docmatix-ir/blob/main/README.md) dataset page.
 Followed by finetuning on the (vidore)[https://huggingface.co/datasets/vidore/colpali_train_set] training set. The checkpoint is warmed up by text retrieval and webpage retrieval.
-For example, DSE-Phi3-Docmatix-V2 achieves **82.9** nDCG@5 on [ViDoRE](https://huggingface.co/spaces/vidore/vidore-leaderboard) leaderboard.
 ## How to train the model from scratch

 The model, `Tevatron/dse-phi35-vidore-ft`, is trained using 1/10 of the `Tevatron/docmatix-ir` dataset, a variant of `HuggingFaceM4/Docmatix` specifically adapted for training PDF retrievers with Vision Language Models in open-domain question answering scenarios. For more information on dataset filtering and hard negative mining, refer to the [docmatix-ir](https://huggingface.co/datasets/Tevatron/docmatix-ir/blob/main/README.md) dataset page.
 Followed by finetuning on the (vidore)[https://huggingface.co/datasets/vidore/colpali_train_set] training set. The checkpoint is warmed up by text retrieval and webpage retrieval.
+For example, DSE-Phi3-Vidore-V2 achieves **82.9** nDCG@5 on [ViDoRE](https://huggingface.co/spaces/vidore/vidore-leaderboard) leaderboard.
 ## How to train the model from scratch