It’s Morphin’ Time! Combating Linguistic Discrimination with Inflectional Perturbations

By Samson Tan, Shafiq R. Joty, Min-Yen Kan, Richard Socher. 2020

121 citations recorded by the source corpus.

Training on only perfect Standard English corpora predisposes pre-trained neural networks to discriminate against minorities from non-standard linguistic backgrounds (e.g., African American Vernacular English, Colloquial Singapore English, etc.). We perturb the inflectional morphology of words to craft plausible and semantically similar adversarial examples that expose these biases in popular NLP models, e.g., BERT and Transformer, and show that adversarially fine-tuning them for a single epoch significantly improves robustness without sacrificing performance on clean data.

Subfields: Artificial Intelligence

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

Citations in the corpus

Language (Technology) is Power: A Critical Survey of “Bias” in NLP (2020, neutral): “…(2019); Sweeney and Najafian (2019); Vaidya et al. (2019); Gaut et al. (2020); Gencoglu (2020); Hovy et al. (2020); Hutchinson et al. (2020); Kim et al. (2020); Peng et al. (2020); Rios (2020); Sap et al. (2020); Shah et al. (2020); Sheng et al. (2020); Tan et al. (2020); Zhang et al. (2020a,b).”
Data Governance in the Age of Large-Scale Data-Driven Language Technology (2022, background): “To this effect, we should be cognizant of the existing linguistic discrimination present in our societies [118] and be careful not to inadvertently replicate them [132, 133].”
A Survey of Race, Racism, and Anti-Racism in NLP (2021, methodology): “Bias Survey/Position van Miltenburg et al. (2017) 2017 INLG Image Processing Detect Bias Waseem et al. (2017) 2017 Workshop Abusive Language Survey/Position Wood-Doughty et al. (2017) 2017 Workshop Social Science/Media Analyze Corpus Wright et al. (2017) 2017 Workshop Abusive Language Analyze Corpus”
NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation (2021, methodology): “This has been adapted from the Morpheus (Tan et al., 2020)”
Quantifying Social Biases in NLP: A Generalization and Empirical Comparison of Extrinsic Fairness Metrics (2021, background): “…group or for each split of a challenge dataset (Hovy and Søgaard, 2015; Rudinger et al., 2018; Zhao et al., 2018; Garimella et al., 2019; Sap et al., 2019; Bagdasaryan et al., 2019; Stafanoviˇcs et al., 2020; Tan et al., 2020; Mehrabi et al., 2020; Nadeem et al., 2020; Cao and Daumé III, 2020).”
Generating Syntactically Controlled Paraphrases without Using Annotated Parallel Pairs (2021, background): “However, the decoder trained on a large corpus also captures the selectional preferences (Katz and Fodor, 1963; Wilks, 1975) in generation, which enables the decoder to infer the proper order of words.”
The practical ethics of bias reduction in machine translation: why domain adaptation is better than data debiasing (2021, background): “More recently, Tan et al. (2020) have exposed NMT models to morphologically varied input English to combat bias effects which reduce the performance for users with non-native linguistic background.”
Better Robustness by More Coverage: Adversarial Training with Mixup Augmentation for Robust Fine-tuning (2020, methodology): “ADA is widely used in the training process to enhance the model robustness (Alzantot et al., 2018; Ren et al., 2019; Zhang et al., 2019; Jin et al., 2020; Li et al., 2020; Tan et al., 2020; Yin et al., 2020; Zheng et al., 2020; Zou et al., 2020; Wang et al., 2020c).”
TextAttack: A Framework for Adversarial Attacks, Data Augmentation, and Adversarial Training in NLP (2020, background): “(Tan et al., 2020) Minimum BLEU Score Inflection Word Swap Greedy search”
Prompting GPT-3 To Be Reliable (2022, unknown): “…et al., 2018; Zhao et al., 2018) and other categories of social biases (Nangia et al., 2020; Nadeem et al., 2021; Parrish et al., 2022), perform poorly against minority demographic groups (Koh et al., 2021; Harris et al., 2022) or dialectical variations (Ziems et al., 2022; Tan et al., 2020).”
CLINE: Contrastive Learning with Semantic Negative Examples for Natural Language Understanding (2021, unknown): “Some works (Alzantot et al., 2018; Tan et al., 2020; Wu et al., 2020) attempt to utilize data augmentation (such as synonym replacement, back translation, etc) to generate positive instances, but few works pay attention to the negative instances.”
Post-hoc Interpretability for Neural NLP: A Survey (2021, background): “Additionally, some works may use adverserial examples to train their model to be more robust (Tan et al., 2020).”