EGA-V2: An End-to-end Generative Framework for Industrial Advertising

TL;DR

EGA-V2 introduces a comprehensive end-to-end generative framework for industrial advertising that models user interests, creative generation, ad allocation, and payment optimization within a single unified system, improving performance over traditional methods.

Contribution

It is the first unified generative model addressing key components of industrial advertising, including bidding, creative selection, and payment, for real-world deployment.

Findings

Outperforms traditional cascaded systems in offline evaluations

Effectively models user interests, POI, and creatives jointly

Ensures incentive compatibility at the POI level

Abstract

Traditional online industrial advertising systems suffer from the limitations of multi-stage cascaded architectures, which often discard high-potential candidates prematurely and distribute decision logic across disconnected modules. While recent generative recommendation approaches provide end-to-end solutions, they fail to address critical advertising requirements of key components for real-world deployment, such as explicit bidding, creative selection, ad allocation, and payment computation. To bridge this gap, we introduce End-to-End Generative Advertising (EGA-V2), the first unified framework that holistically models user interests, point-of-interest (POI) and creative generation, ad allocation, and payment optimization within a single generative model. Our approach employs hierarchical tokenization and multi-token prediction to jointly generate POI recommendations and ad creatives, while a permutation-aware reward model and token-level bidding strategy ensure alignment with both user experiences and advertiser objectives. Additionally, we decouple allocation from payment using a differentiable ex-post regret minimization mechanism, guaranteeing approximate incentive compatibility at the POI level. Through extensive offline evaluations we demonstrate that EGA-V2 significantly outperforms traditional cascaded systems in both performance and practicality. Our results highlight its potential as a pioneering fully generative advertising solution, paving the way for next-generation industrial ad systems.