https://github.com/tencent-ailab/folnet

This repository includes the code for First-Order Logic Network (FOLNet).
https://github.com/tencent-ailab/folnet

Last synced: 2 days ago
JSON representation

This repository includes the code for First-Order Logic Network (FOLNet).

• Host: GitHub
• URL: https://github.com/tencent-ailab/folnet
• Owner: tencent-ailab
• License: apache-2.0
• Created: 2023-09-19T02:08:20.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-09-19T04:06:12.000Z (about 1 year ago)
• Last Synced: 2023-09-19T06:41:38.308Z (about 1 year ago)
• Language: Python
• Size: 84 KB
• Stars: 3
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Disclaimer

This is not an officially supported Tencent product.

# Overview

This repository contains the code for FOLNet (First-Order Logic Network), including the model implementation, the pretraining pipeline and the finetuning pipeline. The full paper on FOLNet is accepted to ICLR2023 with openreview link "[Learning Language Representations with Logical Inductive Bias](https://openreview.net/forum?id=rGeZuBRahju)".

# Docker environment setup

The code can be run by using the docker environment, where the docker image can be built by using the dockerfile in the directory `docker`:

`docker build -t folnet .`

# Usage

## Pretraining

Run pretraining using by using the script `pretrain_FOLNet.sh`.

## Finetuning

1. Finetuning on GLUE: `finetune_FOLNet_GLUE.sh`

2. Finetuning on SQuADv2: `finetune_FOLNet_SQuADv2.sh`

# Model checkpoint

- The model checkpoints can be downloaded from:

[FOLNet-checkpoints](https://tencentoverseas-my.sharepoint.com/:f:/g/personal/jianshuchen_global_tencent_com/Em7QLOIa6bZGuqLoqylyow4BZrL-k3ZiWysXE7tiyyAxjA?e=bnAeH9) or [FOLNet-checkpoints2](https://tencentoverseas-my.sharepoint.com/:f:/g/personal/xiaomanpan_global_tencent_com/EvRCMNsu1NRAlF7Uestta54BeKWJhDk_Gw_RNzdcMAwgSw?e=Kfi4Zw)

- You need to have both the `*.cfg` file (model configuration) and the `*.pt` file (model file).

# Reference

Please cite the paper in the following format if you use this model during your research.

```

@inproceedings{chen2022learning,

  title={Learning Language Representations with Logical Inductive Bias},

  author={Chen, Jianshu},

  booktitle={International Conference on Learning Representations (ICLR)},

  address={Kigali, Rwanda}

  month={May},

  year={2023}

}

```