Mitigating Biases in Toxic Language Detection through Invariant Rationalization.
Yung-Sung ChuangMingye GaoHongyin LuoJames R. GlassHung-Yi LeeYun-Nung ChenShang-Wen LiPublished in: CoRR (2021)
Keyphrases
- false positives
- programming language
- detection algorithm
- detection accuracy
- natural language
- false alarms
- language learning
- detection scheme
- target detection
- affine invariant
- detection method
- object detection
- detection rate
- automatic detection
- language processing
- data sets
- change detection
- database
- affine transformation
- digital images
- modeling language
- feature extraction