Nürnberg NLP 在 ChildSafeAds 2026:四种数据访问级别下的结构相异投票器集成
Nürnberg NLP at ChildSafeAds 2026: Structurally Dissimilar Voter Ensembles under Four Levels of Data Access
- Technische Hochschule Nürnberg Georg Simon Ohm(纽伦堡应用科学大学)
机构由 AI 辅助整理,请以论文原文为准。
AI总结:
针对儿童 YouTube 商业内容监控任务,提出由九个投票器组成的三分支结构相异集成,在通道不相交交叉验证下选择,赢得两个子任务并获任务平均第三。
AI中文摘要:
我们描述了用于 ChildSafeAds 2026 的 Nürnberg NLP 系统。该共享任务探讨的是,在给定数据访问级别下,面向儿童 YouTube 视频的商业内容监控系统能达到何种效果。我们的回答是,针对每个子任务使用由九个投票器组成的集成,这些投票器分为三个分支,分支在骨干网络、适应方法和类别范围上各不相同。选择基于通道不相交的交叉验证,并以开发集作为迁移检查。该系统赢得了三个子任务中的两个。其产品类别得分(ST2,0.8243)和合规标志得分(ST3,0.6530)是 22 个最终参赛作品中最高的,并且在任务平均值(0.7079)上排名第三。我们进一步比较了四种访问级别,并报告了测试集规模下的成本。
英文摘要:
We describe the Nürnberg NLP system for ChildSafeAds 2026. The shared task asks what a monitoring system for commercial content in child-facing YouTube videos can achieve at a given level of data access. We answer with per-subtask ensembles of nine voters, organised into three branches that differ in backbone, adaptation method and class scope. Selection rests on channel-disjoint cross-validation, with the development set as a transfer check. The system wins two of the three subtasks. Its product-category score (ST2, 0.8243) and its compliance-flag score (ST3, 0.6530) are the best of the 22 final entries, and it places third on the task mean (0.7079). We further compare four access levels and report the cost at test-set scale.