Exploring the Relationship Between Diversity and Quality in Ad Text Generation

2025-05-22

AI Summary

Prior work fails to systematically disentangle the relationship between diversity and quality in ad text generation, and general NLP diversity strategies underperform in advertising contexts. Method: We propose the first diversity–quality co-evaluation framework for ad generation, built upon controllable T5/BART/LLaMA-Adapter models and integrating n-gram entropy, BERTScore, CTR simulation, and multi-dimensional human evaluation. Contribution/Results: We empirically uncover a nonlinear trade-off: moderate diversity improves performance, whereas excessive diversity harms relevance—yielding up to a 12.7% CTR gain at optimal diversity. We further introduce AdDiv, a novel diversity-aware quality metric, achieving a Spearman correlation of 0.83 with human quality judgments across three ad datasets. This work provides both theoretical insights and a practical toolset for controllable ad generation.

Technology Category

• Search and Optimization: Evaluation and Analysis
• Natural Language Processing: Generation
• Multiagent Systems: Adversarial Agents

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Ad search and search for Web retail
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

In natural language generation for advertising, creating diverse and engaging ad texts is crucial for capturing a broad audience and avoiding advertising fatigue. Regardless of the importance of diversity, the impact of the diversity-enhancing methods in ad text generation -- mainly tested on tasks such as summarization and machine translation -- has not been thoroughly explored. Ad text generation significantly differs from these tasks owing to the text style and requirements. This research explores the relationship between diversity and ad quality in ad text generation by considering multiple factors, such as diversity-enhancing methods, their hyperparameters, input-output formats, and the models.

Problem

• Exploring diversity-quality relationship in ad text generation
• Assessing diversity methods' impact on ad-specific requirements
• Analyzing factors like hyperparameters, formats, and models

Innovation

• Diversity-enhancing methods in ad text generation
• Exploring hyperparameters and input-output formats
• Assessing relationship between diversity and ad quality