dykang / adventure

code for ACL 2018 paper by Kang et al., "AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples "

Adventure

Code for "Adventure: Adversarial Training for Textual Entailment with Knowledge-Guided Examples" by Dongyeop Kang, Tushar Khot, Ashish Sabharwal and Eduard Hovy, ACL 2018

NOTE:

• For s2s training, you need pre-trained entailment classifier. In this work, we used Decomposable Attention model.
• Maintenance of this repository might be little slow.

Citation

  @inproceedings{kang18acl,
      title = {Adversarial Training for Textual Entailment with Knowledge-Guided Examples},
      author = {Dongyeop Kang and Tushar Khot and Ashish Sabharwal and Eduard Hovy},
      booktitle = {The 56th Annual Meeting of the Association for Computational Linguistics (ACL)},
      address = {Melbourne, Australia},
      month = {July},
      year = {2018}
  }