Summary of Hate Speech Detection with Generalizable Target-aware Fairness, by Tong Chen et al.
Hate Speech Detection with Generalizable Target-aware Fairness
by Tong Chen, Danny Wang, Xurong Liang, Marten Risius, Gianluca Demartini, Hongzhi Yin
First submitted to arxiv on: 28 May 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes Generalizable Target-Aware Fairness (GetFair), a new method for hate speech detection that tackles the issue of bias towards specific targeted groups on social media platforms. Traditional fairness-aware methods are limited to known and fixed target groups, preventing them from generalizing to real-world use cases where new targets emerge. GetFair trains a series of filter functions in an adversarial pipeline to remove the classifier’s dependence on target-related features. The method uses a hypernetwork to generate weights for each target-specific filter without storing dedicated filter parameters. This allows GetFair to achieve advantageous performance on out-of-sample targets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about making sure that online speech detectors don’t unfairly target certain groups of people, like women or black people. Right now, these detectors can get biased towards specific groups and make mistakes. The new method, called GetFair, tries to fix this by removing the detector’s reliance on features related to those groups. It does this by training special filters that “deceive” the detector into thinking it’s seeing different groups of people. This allows GetFair to work better with new or unexpected groups. The paper shows that GetFair performs well even when it encounters groups it hasn’t seen before. |
