HateClipSeg: A Segment-Level Annotated Dataset for Fine-Grained Hate Video Detection

TL;DR

HateClipSeg introduces a large-scale, fine-grained multimodal dataset with segment-level annotations for hate speech detection in videos, enabling more precise benchmarking and model development.

Contribution

The paper presents HateClipSeg, a novel dataset with detailed annotations and benchmarks for hate speech detection in videos, addressing limitations of previous datasets.

Findings

Current models show significant performance gaps.

The dataset enables evaluation of multimodal and temporal detection methods.

High inter-annotator agreement validates annotation quality.

Abstract

Detecting hate speech in videos remains challenging due to the complexity of multimodal content and the lack of fine-grained annotations in existing datasets. We present HateClipSeg, a large-scale multimodal dataset with both video-level and segment-level annotations, comprising over 11,714 segments labeled as Normal or across five Offensive categories: Hateful, Insulting, Sexual, Violence, Self-Harm, along with explicit target victim labels. Our three-stage annotation process yields high inter-annotator agreement (Krippendorff's alpha = 0.817). We propose three tasks to benchmark performance: (1) Trimmed Hateful Video Classification, (2) Temporal Hateful Video Localization, and (3) Online Hateful Video Classification. Results highlight substantial gaps in current models, emphasizing the need for more sophisticated multimodal and temporally aware approaches. The HateClipSeg dataset are publicly available at https://github.com/Social-AI-Studio/HateClipSeg.git.