Counterfactual Probing for the Influence of Affect and Specificity on Intergroup Bias
Not on the map. Why?
Abstract (arXiv)
While existing work on studying bias in NLP focues on negative or pejorative language use, Govindarajan et al. (2023) offer a revised framing of bias in terms of intergroup social context, and its effects on language behavior. In this paper, we investigate if two pragmatic features (specificity and affect) systematically vary in different intergroup contexts -- thus connecting this new framing of bias to language output. Preliminary analysis finds modest correlations between specificity and affect of tweets with supervised intergroup relationship (IGR) labels. Counterfactual probing further reveals that while neural models finetuned for predicting IGR labels reliably use affect in classification, the model's usage of specificity is inconclusive. Code and data can be found at: https://github.com/venkatasg/intergroup-probing
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 13 of its 37 references that are in the map. First posted on arXiv 25 May 2023.
Builds on
13 of 37 references.
References · 13
- 2019The Woman Worked as a Babysitter: On Biases in Language Generationcited by 541
- 2020GoEmotions: A Dataset of Fine-Grained Emotionscited by 486
- 2020Null It Out: Guarding Protected Attributes by Iterative Nullspace Projectioncited by 455
- 2019Social Bias Frames: Reasoning about Social and Power Implications of Languagecited by 380
- 2020BERTweet: A pre-trained language model for English Tweetscited by 346
- 2020Towards Controllable Biases in Language Generationcited by 112
- 2019Automatically Neutralizing Subjective Bias in Textcited by 105
- 2019Gender-preserving Debiasing for Pre-trained Word Embeddingscited by 61
- 2021Counterfactual Interventions Reveal the Causal Effect of Relative Clause Representations on Agreement Predictioncited by 59
- 2022Kernelized Concept Erasurecited by 48
- 2020Detecting Perceived Emotions in Hurricane Disasterscited by 14
- 2020Are "Undocumented Workers" the Same as "Illegal Aliens"? Disentangling Denotation and Connotation in Vector Spacescited by 10
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.