# NeuDEF

**Repository Path**: thunlp/NeuDEF

## Basic Information

- **Project Name**: NeuDEF
- **Description**: The source codes of ICTIR 2019 paper "Neural Document Expansion with User Feedback".
- **Primary Language**: Python
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-05-29
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# NeuDEF

This is the reference implementation of the Neural Document Expansion with User Feedback (NeuDEF) model from paper "Neural Document Expansion with User Feedback". 

## Requirements

- python 2.7
- torch 0.4.1
- numpy 1.16.3

## Efficiency

During training, it takes about 300ms to process one batch on a single-GPU machine with the following settings:

- batch size: 64
- max_q_len: 10
- max_d_len: 50
- max_body_len: 100
- max_exp_len: 10
- max_exp_num: 10
- vocabulary_size: 200K
- embedding dimension: 300
- multi-head attention layer: 1
- multi-head attention head: 4
- learning rate: 0.001

## Results

Please refer our [paper](https://arxiv.org/pdf/1908.02938.pdf).

## Citation

arXiv version:

```
@article{yin2019neural,
  title={Neural Document Expansion with User Feedback},
  author={Yin, Yue and Xiong, Chenyan and Luo, Cheng and Liu, Zhiyuan},
  journal={arXiv preprint arXiv:1908.02938},
  year={2019}
}
```



## Contact

If you have questions, suggestions and bug reports, please email bnuyinyue@outlook.com.