Embedding-based Retrieval in Facebook Search

Abstract

Search in social networks such as Facebook poses different challenges than in classical web search: besides the query text, it is important to take into account the searcher's context to provide relevant results. Their social graph is an integral part of this context and is a unique aspect of Facebook search. While embedding-based retrieval (EBR) has been applied in eb search engines for years, Facebook search was still mainly based on a Boolean matching model. In this paper, we discuss the techniques for applying EBR to a Facebook Search system. We introduce the unified embedding framework developed to model semantic embeddings for personalized search, and the system to serve embedding-based retrieval in a typical search system based on an inverted index. We discuss various tricks and experiences on end-to-end optimization of the whole system, including ANN parameter tuning and full-stack optimization. Finally, we present our progress on two selected advanced topics about modeling. We evaluated EBR on verticals for Facebook Search with significant metrics gains observed in online A/B experiments. We believe this paper will provide useful insights and experiences to help people on developing embedding-based retrieval systems in search engines.

9 pages, 3 figures, 3 tables, to be published in KDD '20

Cited by in corpus (43)

• Semantic Models for the First-stage Retrieval: A Comprehensive Review
• Starling: An I/O-Efficient Disk-Resident Graph Index Framework for High-Dimensional Vector Similarity Search on Data Segment
• Neurosymbolic AI for Reasoning over Knowledge Graphs: A Survey
• TextGNN: Improving Text Encoder via Graph Neural Network in Sponsored Search
• Adap-$\tau$: Adaptively Modulating Embedding Magnitude for Recommendation
• Joint Learning of Deep Retrieval Model and Product Quantization based Embedding Index
• Scenario-Adaptive and Self-Supervised Model for Multi-Scenario Personalized Recommendation
• Revisiting Neural Retrieval on Accelerators
• Mixture of Virtual-Kernel Experts for Multi-Objective User Profile Modeling
• CT4Rec: Simple yet Effective Consistency Training for Sequential Recommendation
• MV-HAN: A Hybrid Attentive Networks based Multi-View Learning Model for Large-scale Contents Recommendation
• STGIN: Spatial-Temporal Graph Interaction Network for Large-scale POI Recommendation
• Unbiased Learning to Rank with Biased Continuous Feedback
• Relevance Filtering for Embedding-based Retrieval
• DARTH: Declarative Recall Through Early Termination for Approximate Nearest Neighbor Search
• LightSAGE: Graph Neural Networks for Large Scale Item Retrieval in Shopee's Advertisement Recommendation
• Learning Multi-Stage Multi-Grained Semantic Embeddings for E-Commerce Search
• OmniSearchSage: Multi-Task Multi-Entity Embeddings for Pinterest Search
• Fast Graph Vector Search via Hardware Acceleration and Delayed-Synchronization Traversal
• Integrity and Junkiness Failure Handling for Embedding-based Retrieval: A Case Study in Social Network Search
• CROLoss: Towards a Customizable Loss for Retrieval Models in Recommender Systems
• An Analysis on Matching Mechanisms and Token Pruning for Late-interaction Models
• HeterRec: Heterogeneous Information Transformer for Scalable Sequential Recommendation
• Heterogeneous Graph Neural Networks for Large-Scale Bid Keyword Matching
• WebANNS: Fast and Efficient Approximate Nearest Neighbor Search in Web Browsers
• Optimizing Context-Enhanced Relational Joins
• Rethinking Large-scale Pre-ranking System: Entire-chain Cross-domain Models
• PERSCEN: Learning Personalized Interaction Pattern and Scenario Preference for Multi-Scenario Matching
• Efficient Optimization Methods for Extreme Similarity Learning with Nonlinear Embeddings
• Prototypical Contrastive Learning and Adaptive Interest Selection for Candidate Generation in Recommendations
• ESANS: Effective and Semantic-Aware Negative Sampling for Large-Scale Retrieval Systems
• CSMF: Cascaded Selective Mask Fine-Tuning for Multi-Objective Embedding-Based Retrieval
• Graph-Based Model-Agnostic Data Subsampling for Recommendation Systems
• Exploring the Upper Limits of Text-Based Collaborative Filtering Using Large Language Models: Discoveries and Insights
• Hessian-aware Quantized Node Embeddings for Recommendation
• A Learnable Fully Interacted Two-Tower Model for Pre-Ranking System
• InteractRank: Personalized Web-Scale Search Pre-Ranking with Cross Interaction Features
• PEFA: Parameter-Free Adapters for Large-scale Embedding-based Retrieval Models
• kNN-Embed: Locally Smoothed Embedding Mixtures For Multi-interest Candidate Retrieval
• SAQ: Pushing the Limits of Vector Quantization through Code Adjustment and Dimension Segmentation
• MPFormer: Adaptive Framework for Industrial Multi-Task Personalized Sequential Retriever
• Embedding Subspace Partitioning for Dynamic Multi-Objective Retrieval
• GRIT: Graph-based Recall Improvement for Task-oriented E-commerce Queries