{"ID":6138224,"CreatedAt":"2026-07-09T01:07:32.349475501Z","UpdatedAt":"2026-07-11T11:54:13.695724419Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.07318","arxiv_id":"2607.07318","title":"R^3: Advertisement Compliance Rectification via Group-Relative Experience Extractor and Curriculum Reinforcement","abstract":"Rigorous content moderation is crucial for online advertising but leads to millions of daily rejections. This scale renders manual rectification infeasible, particularly for video advertisements. However, existing safety-driven methods often suffer from aggressive over-editing, which compromises the advertiser's original semantic intent merely to satisfy compliance. In this work, we target the rectification of textual violations in video ads, covering both speech transcripts and on-screen text. We propose R^3, a novel framework designed to harmonize compliance with original semantic intent preservation. Our approach integrates three key innovations: (1) an experience-driven data synthesis framework that bootstraps high-quality supervision via a group-Relative compliance experience extractor; (2) a curriculum Reinforcement learning strategy with hierarchical rewards designed to enforce compliance while maximizing semantic consistency; and (3) a comprehensive video Rectification framework seamlessly integrating text recognition, rewriting, and re-rendering for industrial deployment. Extensive experiments on industrial datasets and online A/B testing demonstrate that R^3 significantly outperforms state-of-the-art baselines, achieving an optimal trade-off between violation rectification and intent preservation.","short_abstract":"Rigorous content moderation is crucial for online advertising but leads to millions of daily rejections. This scale renders manual rectification infeasible, particularly for video advertisements. However, existing safety-driven methods often suffer from aggressive over-editing, which compromises the advertiser's origin...","url_abs":"https://arxiv.org/abs/2607.07318","url_pdf":"https://arxiv.org/pdf/2607.07318v1","authors":"[\"Yuan Chen\",\"Zhenyu Hu\",\"Mengge Xue\",\"Te Cao\",\"Liqun Liu\",\"Peng Shu\",\"Huan Yu\",\"Jie Jiang\"]","published":"2026-07-08T12:05:41Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.LG\"]","methods":"[\"Reinforcement Learning\"]","has_code":false}