ChildSafeAds Shared Task 2026: Detecting Commercial Content in Child-Facing YouTube Videos

Forum topic · 小凯 · 2026-08-21

Summary

ChildSafeAds is a shared task focused on detecting commercial content in YouTube videos likely to reach children and teenagers. The dataset contains 3,360 videos from 939 channels, with instances anchored on sponsor segments submitted to SponsorBlock, an open-source crowdsourced browser extension whose users mark sponsored portions of videos. Each segment is paired with its transcript, video and channel metadata, and a sales or service page linked from the video description. Participants tackle three subtasks: classifying the type of offer promoted (ST1), assigning product categories (ST2), and identifying legal risk flags (ST3). Evidence is organized into four cumulative access levels, from transcripts to linked pages, allowing performance to be compared against data collection costs. Notably, 45.5% of videos in the dataset fail to correctly use the platform's in-app paid-promotion disclosure label. Labels were generated by GPT-5.4 after expert review of samples and iterative refinement of the taxonomy, prompts, and model selection, with GPT-5.6-luna independently annotating the development set. The report describes the task, data, and evaluation; future versions will add participating systems and final shared task results. Full details are available on arXiv (2608.19165).

Overview

Field: NLP Authors: Thales Bertaglia, Catalina Goanta, Gerasimos Spanakis, Gunes Acar Published: 2026-08-19 arXiv: 2608.19165

Abstract

ChildSafeAds is a shared task on commercial content in YouTube videos likely to reach children and teenagers. It contains 3,360 videos from 939 channels. Each instance begins with a segment submitted to SponsorBlock, an open-source crowdsourced browser extension whose users mark sponsor segments so that others can skip them. The segment is paired with its available transcript, video and channel information, and a sales or service page linked from the video description.

Systems are asked to:

ST1: Determine what kind of offer is being promoted
ST2: Assign product categories
ST3: Identify legal risk flags

The evidence is divided into four cumulative access levels, from the transcript to the linked page, so results can be compared against the cost of collecting the data.

Key Findings

45.5% of videos in the dataset fail to correctly use the platform's in-app ad disclosure method (the "includes paid promotion" label).
Labels were generated by GPT-5.4 after an expert review team examined samples and iteratively refined the taxonomy, prompts, and model selection.
GPT-5.6-luna independently annotated the development set.

This report describes the task, data, and evaluation. Updated versions will add participating systems and shared task results.

---

*Auto-collected on 2026-08-21*