Instructions to use SparseCL/UAE-SparseCL-msmarco with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use SparseCL/UAE-SparseCL-msmarco with Transformers:
# Load model directly from transformers import AutoTokenizer, our_BertForCL tokenizer = AutoTokenizer.from_pretrained("SparseCL/UAE-SparseCL-msmarco") model = our_BertForCL.from_pretrained("SparseCL/UAE-SparseCL-msmarco", device_map="auto") - Notebooks
- Google Colab
- Kaggle
UAE-SparseCL-msmarco
This repository contains a SparseCL model based on WhereIsAI/UAE-Large-V1 and trained on the MS MARCO dataset.
SparseCL is designed for contradiction retrieval.
For implementation and usage details, please see the SparseCL repository.
Citation
Please cite our paper if you use this model:
@inproceedings{xu2025contradiction,
title={Contradiction Retrieval via Contrastive Learning with Sparsity},
author={Xu, Haike and Lin, Zongyu and Chang, Kai-Wei and Sun, Yizhou and Indyk, Piotr},
booktitle={International Conference on Machine Learning},
year={2025}
}
- Downloads last month
- 12
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support