# DeepGen: Diverse Search Ad Generation and Real-Time Customization

arXiv: https://arxiv.org/abs/2208.03438 · HTML: https://mlsanity.com/paper/2208.03438/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 42 references that are in the map.
- First posted on arXiv 6 Aug 2022.

## Builds on (most cited first, 17 of 17)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [On Faithfulness and Factuality in Abstractive Summarization](https://mlsanity.com/paper/2005.00661/) (2020)
- [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://mlsanity.com/paper/1909.05858/) (2019)
- [Unified Language Model Pre-training for Natural Language Understanding and Generation](https://mlsanity.com/paper/1905.03197/) (2019)
- [Texygen: A Benchmarking Platform for Text Generation Models](https://mlsanity.com/paper/1802.01886/) (2018)
- [UniLMv2: Pseudo-Masked Language Models for Unified Language Model Pre-Training](https://mlsanity.com/paper/2002.12804/) (2020)
- [Controlling Linguistic Style Aspects in Neural Language Generation](https://mlsanity.com/paper/1707.02633/) (2017)
- [Evaluating the Evaluation of Diversity in Natural Language Generation](https://mlsanity.com/paper/2004.02990/) (2020)
- [Entity-level Factual Consistency of Abstractive Text Summarization](https://mlsanity.com/paper/2102.09130/) (2021)
- [Controlled Hallucinations: Learning to Generate Faithfully from Noisy Data](https://mlsanity.com/paper/2010.05873/) (2020)
- [Mixture Models for Diverse Machine Translation: Tricks of the Trade](https://mlsanity.com/paper/1902.07816/) (2019)
- [DP-GAN: Diversity-Promoting Generative Adversarial Network for Generating Informative and Diversified Text](https://mlsanity.com/paper/1802.01345/) (2018)
- [Click-Through Rate Prediction in Online Advertising: A Literature Review](https://mlsanity.com/paper/2202.10462/) (2022)
- [Open Domain Web Keyphrase Extraction Beyond Language Modeling](https://mlsanity.com/paper/1911.02671/) (2019)
- [Controllable and Diverse Text Generation in E-commerce](https://mlsanity.com/paper/2102.11497/) (2021)
- [FastSeq: Make Sequence Generation Faster](https://mlsanity.com/paper/2106.04718/) (2021)
- [Automatic Construction of Enterprise Knowledge Base](https://mlsanity.com/paper/2106.15085/) (2021)

## Built on by (most cited first, 10 of 10)

- [Striking Gold in Advertising: Standardization and Exploration of Ad Text Generation](https://mlsanity.com/paper/2309.12030/) (2023)
- [We Should Evaluate Real-World Impact](https://mlsanity.com/paper/2507.05973/) (2025)
- [Natural Language Generation for Advertising: A Survey](https://mlsanity.com/paper/2306.12719/) (2023)
- [FAST: Improving Controllability for Text Generation with Feedback Aware Self-Training](https://mlsanity.com/paper/2210.03167/) (2022)
- [AdTEC: A Unified Benchmark for Evaluating Text Quality in Search Engine Advertising](https://mlsanity.com/paper/2408.05906/) (2024)
- [Aligning the Query Space: Greedy Information Projection for Language Model Data Selection](https://mlsanity.com/paper/2603.13790/) (2026)
- [FaithCAMERA: Construction of a Faithful Dataset for Ad Text Generation](https://mlsanity.com/paper/2410.03839/) (2024)
- [Exploring the Relationship Between Diversity and Quality in Ad Text Generation](https://mlsanity.com/paper/2505.16418/) (2025)
- [Recontextualizing Famous Quotes for Brand Slogan Generation](https://mlsanity.com/paper/2602.06049/) (2026)
- [A Diversity Diet for a Healthier Model: A Case Study of French ModernBERT](https://mlsanity.com/paper/2602.22014/) (2026)

## Abstract (arXiv)

We present DeepGen, a system deployed at web scale for automatically creating sponsored search advertisements (ads) for BingAds customers. We leverage state-of-the-art natural language generation (NLG) models to generate fluent ads from advertiser's web pages in an abstractive fashion and solve practical issues such as factuality and inference speed. In addition, our system creates a customized ad in real-time in response to the user's search query, therefore highlighting different aspects of the same product based on what the user is looking for. To achieve this, our system generates a diverse choice of smaller pieces of the ad ahead of time and, at query time, selects the most relevant ones to be stitched into a complete ad. We improve generation diversity by training a controllable NLG model to generate multiple ads for the same web page highlighting different selling points. Our system design further improves diversity horizontally by first running an ensemble of generation models trained with different objectives and then using a diversity sampling algorithm to pick a diverse subset of generation results for online selection. Experimental results show the effectiveness of our proposed system design. Our system is currently deployed in production, serving ~4% of global ads served in Bing.