SEPP: Similarity Estimation of Predicted Probabilities for Defending and Detecting Adversarial Text
Hoang-Quoc Nguyen-Son, Seira Hidano, Kazuhide Fukushima, Shinsaku Kiyomoto
- Anthology ID:
- 2021.paclic-1.2
- Volume:
- Proceedings of the 35th Pacific Asia Conference on Language, Information and Computation
- Month:
- 11
- Year:
- 2021
- Address:
- Shanghai, China
- Editors:
- Kaibao Hu, Jong-Bok Kim, Chengqing Zong, Emmanuele Chersoni
- Venue:
- PACLIC
- SIG:
- Publisher:
- Association for Computational Lingustics
- Note:
- Pages:
- 11–20
- Language:
- URL:
- https://aclanthology.org/2021.paclic-1.2
- DOI:
- Cite (ACL):
- Hoang-Quoc Nguyen-Son, Seira Hidano, Kazuhide Fukushima, and Shinsaku Kiyomoto. 2021. SEPP: Similarity Estimation of Predicted Probabilities for Defending and Detecting Adversarial Text. In Proceedings of the 35th Pacific Asia Conference on Language, Information and Computation, pages 11–20, Shanghai, China. Association for Computational Lingustics.
- Cite (Informal):
- SEPP: Similarity Estimation of Predicted Probabilities for Defending and Detecting Adversarial Text (Nguyen-Son et al., PACLIC 2021)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2021.paclic-1.2.pdf
- Data
- WiC, Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison