This Reads Like That: Deep Learning for Interpretable Natural Language Processing
Claudio Fanconi, Moritz Vandenhirtz, Severin Husmann, Julia Vogt
Abstract
Prototype learning, a popular machine learning method designed for inherently interpretable decisions, leverages similarities to learned prototypes for classifying new data. While it is mainly applied in computer vision, in this work, we build upon prior research and further explore the extension of prototypical networks to natural language processing. We introduce a learned weighted similarity measure that enhances the similarity computation by focusing on informative dimensions of pre-trained sentence embeddings. Additionally, we propose a post-hoc explainability mechanism that extracts prediction-relevant words from both the prototype and input sentences. Finally, we empirically demonstrate that our proposed method not only improves predictive performance on the AG News and RT Polarity datasets over a previous prototype-based approach, but also improves the faithfulness of explanations compared to rationale-based recurrent convolutions.- Anthology ID:
- 2023.emnlp-main.869
- Volume:
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 14067–14076
- Language:
- URL:
- https://aclanthology.org/2023.emnlp-main.869
- DOI:
- 10.18653/v1/2023.emnlp-main.869
- Cite (ACL):
- Claudio Fanconi, Moritz Vandenhirtz, Severin Husmann, and Julia Vogt. 2023. This Reads Like That: Deep Learning for Interpretable Natural Language Processing. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 14067–14076, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- This Reads Like That: Deep Learning for Interpretable Natural Language Processing (Fanconi et al., EMNLP 2023)
- PDF:
- https://preview.aclanthology.org/corrections-2024-05/2023.emnlp-main.869.pdf