We are inviting submissions to the second workshop on Frontiers of Online Advertising, which will be co-located with ACM EC 2025 in Stanford, CA, USA, on July 10th, 2025.
Submission Deadline (Extended): Thursday, May 22rd Friday, May 30th
Notification of Acceptance: Monday, June 9th
Workshop: Thursday, July 10th
The market for online advertising has witnessed exciting changes in recent years, including new opportunities arising from the integration of generative AI and large-language models, the increased adoption of automation to buy and sell advertisements, new privacy standards, state-of-the-art machine learning prediction methods, and novel experimentation approaches. The “Frontiers of Online Advertising” workshop aims to provide a venue for practitioners and academics to explore and discuss these recent trends.
We welcome submissions related to new frontiers of online advertising, including (but not limited to) topics such as:
Generative AI. New artificial intelligence technologies such as generative AI and large-language models (LLMs) have taken the world by storm. These technologies present a host of new opportunities for advertisers and media platforms alike. We encourage innovative submissions exploring how new technologies could impact advertising markets, e.g., the introduction of new advertising formats and monetization opportunities, using LLMs to improve targeting, creative design, bidding, or auction design. In addition, we invite submissions regarding the monetization of new platforms based on AI-agents.
Automation and Autobidding. Nowadays, most advertising opportunities are sold and bought using auctions and automated bidding algorithms. Automated bidders allow advertisers to increase their yield and navigate the growing complexity of advertising markets. These automated bidders simplify advertisers’ bidding process by asking advertisers for their goals and then procuring advertising slots on their behalf by bidding in different auctions. However, the adoption of automated bidders raises the question of whether existing mechanisms continue to perform well, in particular, due to the interaction of automated bidding agents. We welcome papers exploring the study of robust bidding algorithms, auction design in the presence of automation, and learning in the presence of self-interested agents, among others.
Privacy, Trust, and Regulation. Public awareness regarding data privacy has grown steadily, fueled by high-profile data breaches and increasing concerns about the extent of data collection and tracking. Moreover, stricter data privacy regulations have been enacted by governments, granting users more control over their data and imposing requirements on the collection and usage of personal information. We invite submissions exploring privacy-preserving advertising practices that promote transparency and build trust with users.
Machine learning prediction methods. Online advertising has been fueled by the advancement of machine learning models to predict key performance metrics of ads, such as click-through rates and conversion rates. The interactions with auction mechanisms and bidding systems, together with privacy considerations, bring unique challenges to developing and deploying prediction models. We are open to submissions developing novel approaches to machine learning prediction to improve accuracy, efficiency, interpretability, or privacy guarantees in online advertising.
Experimentation and causal inference. Online advertisers and media platforms routinely experiment to test different creatives or algorithms and establish causal relationships. Traditional experimental designs are not applicable in advertising markets because of interference between units, long-term user learning, or system personalization. We invite submissions exploring novel methodologies for causal inference in advertising markets.
Monetization. While advertising remains one of the most successful ways for internet companies to monetize, alternatives such as subscriptions and tier memberships are promising avenues for monetization. Their adoption, however, introduces novel operational complexities for internet companies. For example, services like Netflix and YouTube, which offer both ad-supported and ad-free tiers at different price points, have to contend with complex market design challenges stemming from different possible levels of ad supply, advertiser demand, and user demand. We invite submissions that compare and contrast different monetization models or explore the interrelated nature of markets that offer multiple such models.
Submission site: https://ec25ad.hotcrp.com/
The preferred submission format is a 2-page abstract. Longer submissions are welcome, but content beyond the first two pages will be read at the reviewers' discretion. Submissions will be evaluated on the relevance to the workshop, the academic merit, and potential for impact. This workshop is non-archival. We welcome the submission of recently-published work or work that is under review elsewhere.
The review process is double blind. Authors must take measures to ensure that their identity is not easily revealed from the submission itself. Authors should refer to their prior work in a neutral manner (i.e., instead of saying “We showed” say “XYZ et al. showed”).
Authors of accepted papers will either give 12-minute oral presentation (for spotlights) or present a poster at the workshop.