PulseAugur
EN
LIVE 08:08:49

MLLMs tested as annotators for harmful YouTube videos with new harm taxonomy

By PulseAugur Editorial · [1 sources] ·

Researchers have developed a comprehensive taxonomy for online harm on video platforms, categorizing it into six types: Information, Hate and harassment, Addictive, Clickbait, Sexual, and Physical harms. They also explored the use of multimodal large language models (MLLMs), specifically GPT-4 Turbo, as alternative annotators for harmful videos. A dataset of 19,422 YouTube videos was annotated using MLLMs, crowdworkers, and trained annotators to evaluate the effectiveness of MLLMs in content moderation and research. AI

IMPACT MLLMs show potential as cost-effective alternatives for content moderation and social science research on video platforms.

RANK_REASON Academic paper detailing a new taxonomy and methodology for harmful content detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

MLLMs tested as annotators for harmful YouTube videos with new harm taxonomy

How we ranked this

Signal score
19 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Academic paper detailing a new taxonomy and methodology for harmful content detection. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, safety
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. arXiv cs.AI TIER_1 English(EN) · Claire Wonjeong Jo, Miki Weso{\l}owska, Magdalena Wojcieszak ·

    Harmful YouTube Video Detection: A Taxonomy of Online Harm and MLLMs as Alternative Annotators

    arXiv:2411.05854v2 Announce Type: replace-cross Abstract: Video platforms, used by billions globally, are criticized for exposing users to harmful content. Yet, detecting online harm remains challenging due to inconsistent approaches to what constitutes harm and limitations of hu…

RELATED ENTITIES

RELATED TOPICS