README

Experimental resources for our IJCAI 2020 Paper "TopicKA: Generating Commonsense Knowledge-Aware Dialogue Responses Towards the Recommended Topic Fact.".

News

• (2020.11.30) We jsut released our code (beta)

Preparation

Our code is based on the Tensorflow (1.14.0, Python3.6).

We reuse some codes/scripts from Tensorflow-NMT.

Tested Environment

ACL2020-ConKADI and this project share the same environment.

     conda create -p ~/envs/conkadi python=3.6
     conda activate ~/envs/conkadi 
     conda install tensorflow-gpu==1.14.0

Datasets

Here we provide the processed datasets. If you need the original datasets, please check the cited papers:

• Chinese Weibo (Baidu Disk code: dgm3 / Google Drive): It is built upon three previous open Chinese Weibo datasets (please see our paper for detail), and we collected commonsense knowledge facts from ConceptIO.

It is worth noting that this "Chinese Weibo" is different from the "Chinese Weibo" of our ACL2020-ConKADI.

In addition, to evaluate the model, you need to download a pre-trained Embeddings:

• Chinese： Tencent AI Lab Embedding Corpus for Chinese Words and Phrases

Training, Inference, and Evaluation

We provide config/scripts files in (/config). 'A > B' means A should be executed and trained before running B.

Please see ACL2020-ConKADI for understanding the following scripts.

• TwoStage: train_classifier.sh, train_twostage.sh

• MultiTask: train_joint.sh or train_round.sh

• Latent Variable: train_joint.sh > train_joint_sample.sh , or train_round.sh > train_round_sample.sh

• Gumbel Softmax: train_joint.sh > train_joint_gumbel.sh , or train_round.sh > train_round_gumbel.sh

Note

• The released code/environment/config may involve some changes compared with the internal version. Meanwhile, due to the randomness of parameters and environment, the experimental results may fluctuate.

• If you use other evaluation scripts, the results may be different. In our paper, we uniformly evaluate models using the scripts in this project.

• The pre-trained embeddings cannot cover all appeared words, and thus we use random embeddings; therefore, in terms of Embed-AVG/EX, therefore, such results will have minor differences if you repeat multiple evaluations.

Citation

If you use our code or data, please kindly cite us in your work.

@inproceedings{ijcai2020-521,
  title     = {TopicKA: Generating Commonsense Knowledge-Aware Dialogue Responses Towards the Recommended Topic Fact},
  author    = {Wu, Sixing and Li, Ying and Zhang, Dawei and Zhou, Yang and Wu, Zhonghai},
  booktitle = {Proceedings of the Twenty-Ninth International Joint Conference on
               Artificial Intelligence, {IJCAI-20}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},             
  editor    = {Christian Bessiere},	
  pages     = {3766--3772},
  year      = {2020},
  month     = {7},
  note      = {Main track}
  doi       = {10.24963/ijcai.2020/521},
  url       = {https://doi.org/10.24963/ijcai.2020/521},
}