Rater Cohesion and Quality from a Vicarious Perspective
Asking people to predict how others with different political views would label content reveals hidden biases and improves data quality for content moderation AI.
Deepak Pandita
•
•
1 min readRead more
Vicarious Offense and Noise Audit of Offensive Speech Classifiers: Unifying Human and Machine Disagreement on What is Offensive
We ran a massive experiment: 9 different AI content moderation systems analyzed 92 million YouTube comments about US politics. The results were shocking—different AI systems …
Tharindu Cyril Weerasooriya
•
•
1 min readVicarious Offense and Noise Audit of Offensive Speech Classifiers
Testing 9 different AI content moderation systems on 92 million YouTube comments reveals wildly inconsistent results, while human annotators show strong political bias—proving that …
Tharindu Cyril Weerasooriya
•
•
1 min read