# Generative Recommendation for Large-Scale Advertising

arXiv: https://arxiv.org/abs/2602.22732 · HTML: https://mlsanity.com/paper/2602.22732/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 44 references that are in the map.
- First posted on arXiv 26 Feb 2026.

## Builds on (most cited first, 29 of 29)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [An Introduction to Variational Autoencoders](https://mlsanity.com/paper/1906.02691/) (2019)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization](https://mlsanity.com/paper/2601.05242/) (2026)
- [Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations](https://mlsanity.com/paper/2402.17152/) (2024)
- [Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation](https://mlsanity.com/paper/2311.09049/) (2023)
- [Learnable Item Tokenization for Generative Recommendation](https://mlsanity.com/paper/2405.07314/) (2024)
- [On-Policy RL Meets Off-Policy Experts: Harmonizing Supervised Fine-Tuning and Reinforcement Learning via Dynamic Weighting](https://mlsanity.com/paper/2508.11408/) (2025)
- [SRFT: A Single-Stage Method with Supervised and Reinforcement Fine-Tuning for Reasoning](https://mlsanity.com/paper/2506.19767/) (2025)
- [Scaling FP8 training to trillion-token LLMs](https://mlsanity.com/paper/2409.12517/) (2024)
- [On Softmax Direct Preference Optimization for Recommendation](https://mlsanity.com/paper/2406.09215/) (2024)
- [Towards a Unified View of Large Language Model Post-Training](https://mlsanity.com/paper/2509.04419/) (2025)
- [TokenRec: Learning to Tokenize ID for LLM-based Generative Recommendation](https://mlsanity.com/paper/2406.10450/) (2024)
- [Generating Long Semantic IDs in Parallel for Recommendation](https://mlsanity.com/paper/2506.05781/) (2025)
- [PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations](https://mlsanity.com/paper/2510.07784/) (2025)
- [OneRec-V2 Technical Report](https://mlsanity.com/paper/2508.20900/) (2025)
- [Large Scale Product Graph Construction for Recommendation in E-commerce](https://mlsanity.com/paper/2010.05525/) (2020)
- [QARM: Quantitative Alignment Multi-Modal Recommendation at Kuaishou](https://mlsanity.com/paper/2411.11739/) (2024)
- [M5Product: Self-harmonized Contrastive Learning for E-commercial Multi-modal Pretraining](https://mlsanity.com/paper/2109.04275/) (2021)
- [CoST: Contrastive Quantization based Semantic Tokenization for Generative Recommendation](https://mlsanity.com/paper/2404.14774/) (2024)
- [OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search](https://mlsanity.com/paper/2509.03236/) (2025)
- [Scaling Law of Large Sequential Recommendation Models](https://mlsanity.com/paper/2311.11351/) (2023)
- [FORGE: Forming Semantic Identifiers for Generative Retrieval in Industrial Datasets](https://mlsanity.com/paper/2509.20904/) (2025)
- [Content-Based Collaborative Generation for Recommender Systems](https://mlsanity.com/paper/2403.18480/) (2024)
- [Generative Retrieval with Semantic Tree-Structured Item Identifiers via Contrastive Learning](https://mlsanity.com/paper/2309.13375/) (2023)
- [MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation](https://mlsanity.com/paper/2508.15281/) (2025)

## Built on by (most cited first, 7 of 7)

- [OneReason Technical Report](https://mlsanity.com/paper/2606.06260/) (2026)
- [UniGD: A Unified Generative-Discriminative Framework for Industrial Retrieval](https://mlsanity.com/paper/2608.03150/) (2026)
- [SAGA: Structure-Attended Generative Action Embedding Model that encodes Multi-Surface User Action Sequences](https://mlsanity.com/paper/2608.15429/) (2026)
- [GenMatch: An End-to-End Generative Matching Framework for Micro-View Order-Dispatching in Ride-Hailing](https://mlsanity.com/paper/2608.19751/) (2026)
- [UniPolicy: Unified Objective-Specific Policies for Generative Search Advertising](https://mlsanity.com/paper/2609.20630/) (2026)
- [OneBid: A Unified Auto-Bidding Foundation Model for Diverse oCPX Advertising Scenarios](https://mlsanity.com/paper/2609.21550/) (2026)
- [Guiding the coarse levels of semantic IDs makes the fine levels learnable](https://mlsanity.com/paper/2609.22227/) (2026)

## Abstract (arXiv)

Generative recommendation has recently attracted widespread attention in industry due to its potential for scaling and stronger model capacity. However, deploying real-time generative recommendation in large-scale advertising requires designs beyond large-language-model (LLM)-style training and serving recipes. We present a production-oriented generative recommender co-designed across architecture, learning, and serving, named GR4AD (Generative Recommendation for ADdvertising). As for tokenization, GR4AD proposes UA-SID (Unified Advertisement Semantic ID) to capture complicated business information. Furthermore, GR4AD introduces LazyAR, a lazy autoregressive decoder that relaxes layer-wise dependencies for short, multi-candidate generation, preserving effectiveness while reducing inference cost, which facilitates scaling under fixed serving budgets. To align optimization with business value, GR4AD employs VSL (Value-Aware Supervised Learning) and proposes RSPO (Ranking-Guided Softmax Preference Optimization), a ranking-aware, list-wise reinforcement learning algorithm that optimizes value-based rewards under list-level metrics for continual online updates. For online inference, we further propose dynamic beam serving, which adapts beam width across generation levels and online load to control compute. Large-scale online A/B tests show up to 4.2% ad revenue improvement over an existing DLRM-based stack, with consistent gains from both model scaling and inference-time scaling. GR4AD has been fully deployed in Kuaishou advertising system with over 400 million users and achieves high-throughput real-time serving.