Infer As You Train: A Symmetric Paradigm of Masked Generative for Click-Through Rate Prediction

TL;DR

This work tackles the training–inference asymmetry in generative CTR models by proposing SGCTR, a symmetric masked generative paradigm. It trains with a discrete diffusion process to learn feature dependencies and employs an iterative inference procedure that refines input features by generating and reweighting masked components, improving robustness to noisy features. Empirical results across four datasets and online A/B tests demonstrate that symmetric inference yields measurable gains over both discriminative and previous generative approaches, including a 2.1% relative CTR uplift in production. Overall, SGCTR shows that fully exploiting generative capabilities during online inference can unlock substantial performance benefits for CTR prediction.

Abstract

Generative models are increasingly being explored in click-through rate (CTR) prediction field to overcome the limitations of the conventional discriminative paradigm, which rely on a simple binary classification objective. However, existing generative models typically confine the generative paradigm to the training phase, primarily for representation learning. During online inference, they revert to a standard discriminative paradigm, failing to leverage their powerful generative capabilities to further improve prediction accuracy. This fundamental asymmetry between the training and inference phases prevents the generative paradigm from realizing its full potential. To address this limitation, we propose the Symmetric Masked Generative Paradigm for CTR prediction (SGCTR), a novel framework that establishes symmetry between the training and inference phases. Specifically, after acquiring generative capabilities by learning feature dependencies during training, SGCTR applies the generative capabilities during online inference to iteratively redefine the features of input samples, which mitigates the impact of noisy features and enhances prediction accuracy. Extensive experiments validate the superiority of SGCTR, demonstrating that applying the generative paradigm symmetrically across both training and inference significantly unlocks its power in CTR prediction.

Figures (1)

• Figure 1: Prediction performance under different steps.