Detect and Perturb: Neutral Rewriting of Biased and Sensitive Text via Gradient-based Decoding
By Zexue He, Bodhisattwa Prasad Majumder, Julian McAuley. 2021
34 citations recorded by the source corpus.
Written language carries explicit and implicit biases that can distract from meaningful signals. For example, letters of reference may describe male and female candidates differently, or their writing style may indirectly reveal demographic characteristics. At best, such biases distract from the meaningful content of the text; at worst they can lead to unfair outcomes. We investigate the challenge of re-generating input sentences to 'neutralize' sensitive attributes while maintaining the semantic meaning of the original text (e.g. is the candidate qualified?). We propose a gradient-based rewriting framework, Detect and Perturb to Neutralize (DEPEN), that first detects sensitive components and masks them for regeneration, then perturbs the generation model at decoding time under a neutralizing constraint that pushes the (predicted) distribution of sensitive attributes towards a uniform distribution. Our experiments in two different scenarios show that DEPEN can regenerate fluent alternatives that are neutral in the sensitive attribute while maintaining the semantics of other attributes.
Subfields: Artificial Intelligence, Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Yulia Tsvetkov
- Vidhisha Balachandran
- Sachin Kumar
- Antonios Anastasopoulos
- Lucille Njoo
- Mohit Bansal
- Anjalie Field
- Hinrich Schütze
In Natural Language Processing
- Yulia Tsvetkov
- Sachin Kumar
- Vidhisha Balachandran
- Antonios Anastasopoulos
- Lucille Njoo
- Anjalie Field
- Hinrich Schütze
- Mohit Bansal
Citations in the corpus
- Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey (2022, background): “The first set of works treat the task of rewriting as a sequence labeling task, where each token in the output sequence is either substituted, deleted, or kept the same (Pryzant et al., 2020; He et al., 2021b).”