# TwHIN: Embedding the Twitter Heterogeneous Information Network for Personalized Recommendation

arXiv: https://arxiv.org/abs/2202.05387 · HTML: https://mlsanity.com/paper/2202.05387/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Feb 2022.

## Built on by (most cited first, 25 of 25)

- [Graph Neural Networks for Link Prediction with Subgraph Sketching](https://mlsanity.com/paper/2209.15486/) (2022)
- [TwHIN-BERT: A Socially-Enriched Pre-trained Language Model for Multilingual Tweet Representations at Twitter](https://mlsanity.com/paper/2209.07562/) (2022)
- [HiGPT: Heterogeneous Graph Language Model](https://mlsanity.com/paper/2402.16024/) (2024)
- [Heterformer: Transformer-based Deep Node Representation Learning on Heterogeneous Text-Rich Networks](https://mlsanity.com/paper/2205.10282/) (2022)
- [Scaling User Modeling: Large-scale Online User Representations for Ads Personalization in Meta](https://mlsanity.com/paper/2311.09544/) (2023)
- [Towards Graph Foundation Models for Personalization](https://mlsanity.com/paper/2403.07478/) (2024)
- [A Simulation System Towards Solving Societal-Scale Manipulation](https://mlsanity.com/paper/2410.13915/) (2024)
- [LinkSAGE: Optimizing Job Matching Using Graph Neural Networks](https://mlsanity.com/paper/2402.13430/) (2024)
- [Multi-Scale Heterogeneous Text-Attributed Graph Datasets From Diverse Domains](https://mlsanity.com/paper/2412.08937/) (2024)
- [kNN-Embed: Locally Smoothed Embedding Mixtures For Multi-interest Candidate Retrieval](https://mlsanity.com/paper/2205.06205/) (2022)
- [data2lang2vec: Data Driven Typological Features Completion](https://mlsanity.com/paper/2409.17373/) (2024)
- [NTULM: Enriching Social Media Text Representations with Non-Textual Units](https://mlsanity.com/paper/2210.16586/) (2022)
- [TwERC: High Performance Ensembled Candidate Generation for Ads Recommendation at Twitter](https://mlsanity.com/paper/2302.13915/) (2023)
- [Geometric instability of graph neural networks on large graphs](https://mlsanity.com/paper/2308.10099/) (2023)
- [A Scalable and Efficient Signal Integration System for Job Matching](https://mlsanity.com/paper/2507.09797/) (2025)
- [When Embedding Models Meet: Procrustes Bounds and Applications](https://mlsanity.com/paper/2510.13406/) (2025)
- [Graph-enhanced Optimizers for Structure-aware Recommendation Embedding Evolution](https://mlsanity.com/paper/2310.03032/) (2023)
- [GLINKX: A Scalable Unified Framework For Homophilous and Heterophilous Graphs](https://mlsanity.com/paper/2211.00550/) (2022)
- [URL-BERT: Training Webpage Representations via Social Media Engagements](https://mlsanity.com/paper/2310.16303/) (2023)
- [Understanding the Design Principles of Link Prediction in Directed Settings](https://mlsanity.com/paper/2502.15008/) (2025)
- [Breaker: Removing Shortcut Cues with User Clustering for Single-slot Recommendation System](https://mlsanity.com/paper/2506.00828/) (2025)
- [Who Is Lagging Behind: Profiling Student Behaviors with Graph-Level Encoding in Curriculum-Based Online Learning Systems](https://mlsanity.com/paper/2508.18925/) (2025)
- [Decoupled Entity Representation Learning for Pinterest Ads Ranking](https://mlsanity.com/paper/2509.04337/) (2025)
- [Recommender system in X inadvertently profiles ideological positions of users](https://mlsanity.com/paper/2602.02624/) (2026)
- [The Like Trap: Multi-Stage Poisoning against Agents in Similarity-based Recommendation Systems](https://mlsanity.com/paper/2609.27155/) (2026)

## Abstract (arXiv)

Social networks, such as Twitter, form a heterogeneous information network (HIN) where nodes represent domain entities (e.g., user, content, advertiser, etc.) and edges represent one of many entity interactions (e.g, a user re-sharing content or "following" another). Interactions from multiple relation types can encode valuable information about social network entities not fully captured by a single relation; for instance, a user's preference for accounts to follow may depend on both user-content engagement interactions and the other users they follow. In this work, we investigate knowledge-graph embeddings for entities in the Twitter HIN (TwHIN); we show that these pretrained representations yield significant offline and online improvement for a diverse range of downstream recommendation and classification tasks: personalized ads rankings, account follow-recommendation, offensive content detection, and search ranking. We discuss design choices and practical challenges of deploying industry-scale HIN embeddings, including compressing them to reduce end-to-end model latency and handling parameter drift across versions.