DGenCTR: Towards a Universal Generative Paradigm for Click-Through Rate Prediction via Discrete Diffusion

2025/08/20 by Moyu Zhang, Yun Chen, Zhang, Moyu +7 · 2 citations
Computer Science · Social Sciences · #Advanced Computing and Algorithms #FOS: Computer and information sciences #Image and Video Quality Assessment #Information Retrieval (cs.IR)

paper · pdf · doi:10.48550/arxiv.2508.14500

openalex publication_date 2025/08/20 · openalex created_date 2025/10/16 · openalex updated_date 2026/07/28

Abstract

Recent advances in generative models have inspired the field of recommender systems to explore generative approaches, but most existing research focuses on sequence generation, a paradigm ill-suited for click-through rate (CTR) prediction. CTR models critically depend on a large number of cross-features between the target item and the user to estimate the probability of clicking on the item, and discarding these cross-features will significantly impair model performance. Therefore, to harness the ability of generative models to understand data distributions and thereby alleviate the constraints of traditional discriminative models in label-scarce space, diverging from the item-generation paradigm of sequence generation methods, we propose a novel sample-level generation paradigm specifically designed for the CTR task: a two-stage Discrete Diffusion-Based Generative CTR training framework (DGenCTR). This two-stage framework comprises a diffusion-based generative pre-training stage and a CTR-targeted supervised fine-tuning stage for CTR. Finally, extensive offline experiments and online A/B testing conclusively validate the effectiveness of our framework.

Citations

• MTGR: Industrial-Scale Generative Recommendation Framework in Meituan
• Sparse Meets Dense: Unified Generative Recommendations with Cascaded Sparse-Dense Representations
• OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment
• Fusion Matters: Learning Fusion in Deep Click-through Rate Prediction Models
• Heavy-Tailed Diffusion with Denoising Lévy Probabilistic Models
• EAGER: Two-Stream Generative Recommender with Behavior-Semantic Collaboration
• Learnable Item Tokenization for Generative Recommendation
• IDGenRec: LLM-RecSys Alignment with Textual ID Learning
• Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation
• Towards Deeper, Lighter and Interpretable Cross Network for CTR Prediction
• On the Embedding Collapse when Scaling up Recommendation Models
• A Bi-Step Grounding Paradigm for Large Language Models in Recommendation Systems
• GenRec: Large Language Model for Generative Recommendation
• PEPNet: Parameter and Embedding Personalized Network for Infusing with Personalized Prior Information
• Score-based Continuous-time Discrete Diffusion Models
• Joint Optimization of Ranking and Calibration with Contextualized Hybrid Model
• A Continuous Time Framework for Discrete Denoising Models
• Re4: Learning to Re-contrast, Re-attend, Re-construct for Multi-interest Recommendation
• Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5)
• Alleviating Cold-start Problem in CTR Prediction with A Variational Embedding Learning Framework
• Deep Learning for Click-Through Rate Estimation
• Hierarchically Modeling Micro and Macro Behaviors via Multi-Task Learning for Conversion Rate Prediction
• MaskNet: Introducing Feature-Wise Multiplication to CTR Ranking Models by Instance-Guided Mask
• Empowering Things with Intelligence: A Survey of the Progress, Challenges, and Opportunities in Artificial Intelligence of Things
• Search-based User Interest Modeling with Lifelong Sequential Behavior Data for Click-Through Rate Prediction
• Scaling Laws for Neural Language Models
• Deep Interest Evolution Network for Click-Through Rate Prediction
• Attention Is All You Need
• DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
• Product-based Neural Networks for User Response Prediction
• Wide & Deep Learning for Recommender Systems
• TensorFlow: A system for large-scale machine learning
• Adam: A Method for Stochastic Optimization
• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Cited by

• Infer As You Train: A Symmetric Paradigm of Masked Generative for Click-Through Rate Prediction
• MATT-CTR: Unleashing a Model-Agnostic Test-Time Paradigm for CTR Prediction with Confidence-Guided Inference Paths

Related

• Generative Click-through Rate Prediction with Applications to Search Advertising
• Infer As You Train: A Symmetric Paradigm of Masked Generative for Click-Through Rate Prediction
• From Feature Interaction to Feature Generation: A Generative Paradigm of CTR Prediction Models
• Addressing Cold-start Problem in Click-Through Rate Prediction via Supervised Diffusion Modeling
• Memorize, Factorize, or be Naïve: Learning Optimal Feature Interaction Methods for CTR Prediction