Deep Learning to Rank in Industrial Search Engines, Recommender Systems and Online Advertising: An Overview and New Perspectives

Search engines, Recommender systems and Online advertising are playing fundamental roles in modern web and mobile applications. In these information systems, the most significant component is the ranking system, which selects a list of items likely to interest a user from billions of candidate items. At its core, Deep learning to rank (DLTR) has become indispensable for building high-performance ranking models, driving significant gains in user engagement and business growth. In this paper, firstly, we outline the key problems and challenges in industrial-scale ranking systems. Secondly, we provide a comprehensive review of deep learning models deployed across multiple stages of the industrial ranking pipeline, including matching, pre-ranking, fine-grained ranking, post-ranking, and relevance-ranking. Finally, we explore novel perspectives for future research, such as leveraging Large Language Models (LLMs). The papers discussed in this survey are listed in https://github.com/guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising.

• Publication year

  2026

• Venue

  ACM Transactions on Information Systems

• Publication date

  2026-02-16

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1145/3797895
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• HyFormer: Revisiting the Roles of Sequence Modeling and Feature Interaction in CTR Prediction

  2026cited by this paper
• Qwen3 Technical Report

  2025cited by this paper
• Make It Long, Keep It Fast: End-to-End 10k-Sequence Modeling at Billion Scale on Douyin

  2025influential reference
• Large Vison-Language Foundation Model in Baidu AIGC Image Advertising

  2025cited by this paper
• PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations

  2025influential reference
• FIM: Frequency-Aware Multi-View Interest Modeling for Local-Life Service Recommendation

  2025cited by this paper
• PinFM: Foundation Model for User Activity Sequences at a Billion-scale Visual Discovery Platform

  2025cited by this paper
• OneSug: The Unified End-to-End Generative Framework for E-commerce Query Suggestion

  2025influential reference
• Applying Large Language Model For Relevance Search In Tencent

  2025influential reference
• Real-time Indexing for Large-scale Recommendation by Streaming Vector Quantization Retriever

  2025influential reference
• Sparse Meets Dense: Unified Generative Recommendations with Cascaded Sparse-Dense Representations

  2025influential reference
• MOON: Generative MLLM-based Multimodal Representation Learning for E-commerce Product Understanding

  2025cited by this paper
• PinRec: Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems

  2025influential reference
• LONGER: Scaling Up Long Sequence Modeling in Industrial Recommenders

  2025influential reference
• MUSE: A Simple Yet Effective Multimodal Search-Based Framework for Lifelong User Interest Modeling

  2025cited by this paper
• HCMRM: A High-Consistency Multimodal Relevance Model for Search Ads

  2025influential reference
• HIT Model: A Hierarchical Interaction-Enhanced Two-Tower Model for Pre-Ranking Systems

  2025cited by this paper
• OneLoc: Geo-Aware Generative Recommender Systems for Local Life Service

  2025cited by this paper
• GPR: Towards a Generative Pre-trained One-Model Paradigm for Large-Scale Advertising Recommendation

  2025cited by this paper
• Generative Retrieval and Alignment Model: A New Paradigm for E-commerce Retrieval

  2025influential reference
• OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment

  2025influential reference
• RankMixer: Scaling Up Ranking Models in Industrial Recommenders

  2025influential reference
• OneTrans: Unified Feature Interaction and Sequence Modeling with One Transformer in Industrial Recommender

  2025cited by this paper
• Unlocking Scaling Law in Industrial Recommendation Systems with a Three-step Paradigm based Large User Model

  2025influential reference
• Qwen2.5-VL Technical Report

  2025influential reference
• TransAct V2: Lifelong User Action Sequence Modeling on Pinterest Recommendation

  2025cited by this paper
• A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models

  2024cited by this paper
• Large Language Models for Relevance Judgment in Product Search

  2024influential reference
• Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application

  2024influential reference
• NoteLLM: A Retrievable Large Language Model for Note Recommendation

  2024influential reference
• TWIN V2: Scaling Ultra-Long User Behavior Sequence Modeling for Enhanced CTR Prediction at Kuaishou

  2024influential reference
• NoteLLM-2: Multimodal Large Representation Models for Recommendation

  2024influential reference
• Understanding the Ranking Loss for Recommendation with Sparse User Feedback

  2024cited by this paper
• Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations

  2024influential reference
• Ads Recommendation in a Collapsed and Entangled World

  2024cited by this paper
• From Matching to Generation: A Survey on Generative Information Retrieval

  2024cited by this paper
• MLoRA: Multi-Domain Low-Rank Adaptive Network for CTR Prediction

  2024cited by this paper
• Trinity: Syncretizing Multi-/Long-Tail/Long-Term Interests All in One

  2024influential reference
• Explainable LLM-driven Multi-dimensional Distillation for E-Commerce Relevance Learning

  2024influential reference
• Beyond Binary Preference: Leveraging Bayesian Approaches for Joint Optimization of Ranking and Calibration

  2024cited by this paper
• HLLM: Enhancing Sequential Recommendations via Hierarchical Large Language Models for Item and User Modeling

  2024influential reference
• Wukong: Towards a Scaling Law for Large-Scale Recommendation

  2024influential reference
• HoME: Hierarchy of Multi-Gate Experts for Multi-Task Learning at Kuaishou

  2024influential reference
• Deep Bag-of-Words Model: An Efficient and Interpretable Relevance Architecture for Chinese E-Commerce

  2024cited by this paper
• Non-autoregressive Generative Models for Reranking Recommendation

  2024cited by this paper
• KuaiFormer: Transformer-Based Retrieval at Kuaishou

  2024influential reference
• Enhancing Taobao Display Advertising with Multimodal Representations: Challenges, Approaches and Insights

  2024influential reference
• Recommender Systems with Generative Retrieval

  2023cited by this paper
• A survey on large language models for recommendation

  2023cited by this paper
• Rethinking the Role of Pre-ranking in Large-scale E-Commerce Searching System

  2023influential reference
• Entire Space Learning Framework: Unbias Conversion Rate Prediction in Full Stages of Recommender System

  2023cited by this paper
• TransAct: Transformer-based Realtime User Action Model for Recommendation at Pinterest

  2023influential reference
• GPT-4 Technical Report

  2023influential reference
• Deep reinforcement learning in recommender systems: A survey and new perspectives

  2023cited by this paper
• PEPNet: Parameter and Embedding Personalized Network for Infusing with Personalized Prior Information

  2023influential reference
• TWIN: TWo-stage Interest Network for Lifelong User Behavior Modeling in CTR Prediction at Kuaishou

  2023cited by this paper
• Generative Slate Recommendation with Reinforcement Learning

  2023cited by this paper
• LLaMA: Open and Efficient Foundation Language Models

  2023cited by this paper
• PIER: Permutation-Level Interest-Based End-to-End Re-ranking Framework in E-commerce

  2023cited by this paper
• Query-dominant User Interest Network for Large-Scale Search Ranking

  2023cited by this paper
• Towards Deeper, Lighter and Interpretable Cross Network for CTR Prediction

  2023cited by this paper
• AdaTT: Adaptive Task-to-Task Fusion Network for Multitask Learning in Recommendations

  2023cited by this paper
• Hiformer: Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems

  2023influential reference
• Qwen Technical Report

  2023cited by this paper
• COPR: Consistency-Oriented Pre-Ranking for Online Advertising

  2023influential reference
• RankFormer: Listwise Learning-to-Rank Using Listwide Labels

  2023cited by this paper
• SPM: Structured Pretraining and Matching Architectures for Relevance Modeling in Meituan Search

  2023cited by this paper
• FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

  2022cited by this paper
• Scale Calibration of Deep Ranking Models

  2022influential reference
• APG: Adaptive Parameter Generation Network for Click-Through Rate Prediction

  2022cited by this paper
• ESCM2: Entire Space Counterfactual Multi-Task Model for Post-Click Conversion Rate Estimation

  2022cited by this paper
• Sampling Is All You Need on Modeling Long-Term User Behaviors for CTR Prediction

  2022cited by this paper
• Multi-Behavior Hypergraph-Enhanced Transformer for Sequential Recommendation

  2022cited by this paper
• Asymptotically Unbiased Estimation for Delayed Feedback Modeling via Label Correction

  2022cited by this paper
• Multi-Objective Personalized Product Retrieval in Taobao Search

  2022influential reference
• Modeling Users' Contextualized Page-wise Feedback for Click-Through Rate Prediction in E-commerce Search

  2022cited by this paper
• CAN: Feature Co-Action Network for Click-Through Rate Prediction

  2022cited by this paper
• MemoNet: Memorizing All Cross Features' Representations Efficiently via Multi-Hash Codebook Network for CTR Prediction

  2022cited by this paper
• Approximate Nearest Neighbor Search under Neural Similarity Metric for Large-Scale Recommendation

  2022influential reference
• Joint Optimization of Ranking and Calibration with Contextualized Hybrid Model

  2022influential reference
• On the Factory Floor: ML Engineering for Industrial-Scale Ads Recommendation Models

  2022cited by this paper
• Training language models to follow instructions with human feedback

  2022influential reference
• Filter-enhanced MLP is All You Need for Sequential Recommendation

  2022cited by this paper
• Regression Compatible Listwise Objectives for Calibrated Ranking with Binary Relevance

  2022cited by this paper
• ReprBERT: Distilling BERT to an Efficient Representation-Based Relevance Model for E-Commerce

  2022influential reference
• Path-based Deep Network for Candidate Item Matching in Recommenders

  2021influential reference
• Real Negatives Matter: Continuous Training with Real Negatives for Delayed Feedback Modeling

  2021cited by this paper
• Revisit Recommender System in the Permutation Prospective

  2021cited by this paper
• Learning a Product Relevance Model from Click-Through Data in E-Commerce

  2021influential reference
• Optimizing Dense Retrieval Model Training with Hard Negatives

  2021cited by this paper
• Pre-trained Language Model based Ranking in Baidu Search

  2021cited by this paper
• MLP-Mixer: An all-MLP Architecture for Vision

  2021cited by this paper
• A Dual Augmented Two-tower Model for Online Large-scale Recommendation

  2021cited by this paper
• Handling many conversions per click in modeling delayed feedback

  2021cited by this paper
• Learning Transferable Visual Models From Natural Language Supervision

  2021cited by this paper
• POSO: Personalized Cold Start Modules for Large-scale Recommender Systems

  2021influential reference
• Highly accurate protein structure prediction with AlphaFold

  2021cited by this paper
• Self-Supervised Learning on Users' Spontaneous Behaviors for Multi-Scenario Ranking in E-commerce

  2021influential reference
• Pre-trained Language Model for Web-scale Retrieval in Baidu Search

  2021influential reference
• Attentive Neural Point Processes for Event Forecasting

  2021cited by this paper

• No citing papers are available for this paper.