license: apache-2.0
language:
- zh
- en
- es
- fr
- de
- ru
- ja
- ko
base_model:
- Qwen/Qwen3-Embedding-4B
library_name: transformers
Querit-Reranker-4B
HighLights
Querit-Reranker-4B is a cross-encoder reranking model developed by the Querit family for multilingual text ranking tasks. The model is initialized from Qwen3-Embedding-4B and further adapted for reranking through task-oriented post-training. By jointly encoding each query-document pair, Querit-Reranker-4B captures fine-grained relevance signals and produces accurate ranking scores for second-stage retrieval. With multilingual supervision, teacher-score distillation, and task-specific checkpoint merging, the model achieves strong performance across multilingual, English, and Chinese reranking benchmarks.
Model Description
- Model type: Text Reranking
- Language(s) (NLP): Multilingual (Chinese, English, Spanish, French, German, Russian, Korean, Japanese)
- Training Stage: Pretraining & Post-training
- Number of Total Parameters: 4.02B
- Number of Paramaters (Non-Embedding): 3.63B
- Number of Layers: 36
- Number of Attention Heads: 32
- Context Length: 128k
Citation
If you find Querit-Reranker useful for your research or applications, please cite our paper:
Querit-Reranker: Training Compact Multilingual Rerankers via Efficient Label-Free Distribution Adaptation Yunfei Zhong, Jun Yang, Wei Huang, Yinqiong Cai, Haosheng Qian, Yixing Fan, Ruqing Zhang, Lixin Su, Daiting Shi, and Jiafeng Guo. arXiv:2606.19037, 2026.
@misc{zhong2026queritrerankertrainingcompactmultilingual,
title={Querit-Reranker: Training Compact Multilingual Rerankers via Efficient Label-Free Distribution Adaptation},
author={Yunfei Zhong and Jun Yang and Wei Huang and Yinqiong Cai and Haosheng Qian and Yixing Fan and Ruqing Zhang and Lixin Su and Daiting Shi and Jiafeng Guo},
year={2026},
eprint={2606.19037},
archivePrefix={arXiv},
primaryClass={cs.IR},
url={https://arxiv.org/abs/2606.19037},
}