Using Large Language Models to extract rationales for hate speech detection enables interpretable classifiers.
The author proposes the Attributive Graph Enhanced Embedding (AGEE) algorithm to improve link prediction accuracy by leveraging node attribute information. By optimizing the trade-off between structure and attribute networks, AGEE enhances existing link prediction algorithms.
This study investigates the impact of Opinion Polarization (OP) on short video consumption using EEG signals and sentiment analysis. The authors reveal significant effects on user sentiments, brain activity, and behavior metrics.
The author explores how VTubers use virtual avatars to present exaggerated gender expressions and engage with viewers in inclusive interactions.
The authors explore the challenges faced in restructuring community-based moderation, focusing on the impact of Flagged Revisions on Wikipedia. They highlight conflicts with social norms, unclear moderation instructions, and the struggle to reach consensus as key issues.