用可控前缀提升广告标题生成效果，支持多格式适配
COBART: Controlled, Optimized, Bidirectional and Auto-Regressive Transformer for Ad Headline Generation
- 引入前缀控制令牌，结合BART微调实现精准生成
- 相比基线，ROUGE-L提升25.82%，预估点击率提高5.82%
- 灵活适配不同广告形式与优化目标，适合工业级应用
在线广告对各类企业至关重要，广告标题是其核心创意组件。现有方法虽可自动生标题并优化点击率（CTR）与质量，但不断变化的广告格式和创作需求使生成优化且定制化的标题仍具挑战。本文提出一种新方法：通过在BART微调中引入前缀控制令牌，实现标题生成的精确控制。该方法不仅获得最高点击率，还支持用户设定生成标题长度，以适应不同广告格式。同时具备良好灵活性，可轻松拓展至其他架构、创作需求与优化标准。实验表明，相较已有强基线，该方法在ROUGE-L上提升25.82%，预估点击率提高5.82%。
原文摘要 · Abstract (English)
Online ads are essential to all businesses and ad headlines are one of their core creative component. Existing methods can generate headlines automatically and also optimize their click-through-rate (CTR) and quality. However, evolving ad formats and changing creative requirements make it difficult to generate optimized & customized headlines. We propose a novel method that uses prefix control tokens along with BART fine-tuning. It yields the highest CTR and also allows users to control the length of generated headlines for use across different ad formats. The method is also flexible and can easily be adapted to other architectures, creative requirements and optimization criteria. Our experiments demonstrate a 25.82% increment in Rouge-L and a 5.82% increment in estimated CTR over previously published strong ad headline generation baseline.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。