ML Anthology

Improving Interpretability via Explicit Word Interaction Graph Layer

Arshdeep Sekhon, Hanjie Chen, Aman Shrivastava, Zhe Wang, Yangfeng Ji, Yanjun Qi
AAAI 2023 pp. 13528-13537
doi:10.1609/AAAI.V37I11.26586 /aaai/2023/sekhon2023aaai-improving/

Abstract

Recent NLP literature has seen growing interest in improving model interpretability. Along this direction, we propose a trainable neural network layer that learns a global interaction graph between words and then selects more informative words using the learned word interactions. Our layer, we call WIGRAPH, can plug into any neural network-based NLP text classifiers right after its word embedding layer. Across multiple SOTA NLP models and various NLP datasets, we demonstrate that adding the WIGRAPH layer substantially improves NLP models' interpretability and enhances models' prediction performance at the same time.

PDF AAAI Semantic Scholar

Cite

Text

Sekhon et al. "Improving Interpretability via Explicit Word Interaction Graph Layer." AAAI Conference on Artificial Intelligence, 2023. doi:10.1609/AAAI.V37I11.26586

Markdown

[Sekhon et al. "Improving Interpretability via Explicit Word Interaction Graph Layer." AAAI Conference on Artificial Intelligence, 2023.](https://mlanthology.org/aaai/2023/sekhon2023aaai-improving/) doi:10.1609/AAAI.V37I11.26586

BibTeX

@inproceedings{sekhon2023aaai-improving,
  title     = {{Improving Interpretability via Explicit Word Interaction Graph Layer}},
  author    = {Sekhon, Arshdeep and Chen, Hanjie and Shrivastava, Aman and Wang, Zhe and Ji, Yangfeng and Qi, Yanjun},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2023},
  pages     = {13528-13537},
  doi       = {10.1609/AAAI.V37I11.26586},
  url       = {https://mlanthology.org/aaai/2023/sekhon2023aaai-improving/}
}