ConnectionMind: Leveraging Social Networks and Large Language Models for Personalized Recommendation at Meta

2026-08-10

AI Summary

Traditional recommender systems struggle to integrate complex social relationships and heterogeneous content, lacking fine-grained, personalized reasoning over multi-source contextual signals. This work proposes a novel social-aware recommendation framework that, for the first time in an industrial-scale system, leverages large language models (LLMs) for path-based reasoning on social heterogeneous graphs. The approach employs a two-stage learning strategy—supervised fine-tuning followed by reinforcement learning—to enable scalable, interpretable, end-to-end recommendation. Deployed within Meta’s large-scale recommendation pipeline, the method demonstrates significant real-world impact: online A/B tests reveal a 0.43% increase in video watch time, highlighting its effectiveness in enhancing user engagement through socially informed, contextually rich personalization.

Abstract

Modern recommendation systems on social media platforms such as Meta must model complex social relationships, including friendships, group memberships, and creator interactions, alongside massive and heterogeneous content such as text and video. Traditional recommendation models, however, often omit these signals or treat them independently, lacking the reasoning capability to integrate multi-relational context for fine-grained personalization. We present ConnectionMind, a production-ready recommendation framework that tightly integrates the social network structure with large language models (LLMs) to enable scalable, interpretable, and reasoning-aware personalization in Meta. ConnectionMind constructs a heterogeneous graph connecting users, items, friends, groups, and creator pages, and formulates recommendation as a graph reasoning problem: discovering personalized paths from users to candidate items. An LLM-based policy is employed to reason over these graph structures and guide recommendation decisions. To train the system at scale, ConnectionMind adopts a two-stage learning strategy. We first perform supervised fine-tuning (SFT) on large-scale user-item interaction trajectories to initialize the reasoning policy, followed by end-to-end reinforcement learning (RL) to refine the model's ability to reason over social graphs for personalized recommendation. Extensive experiments on multiple real-world datasets demonstrate the effectiveness of ConnectionMind compared to representative baselines. More importantly, ConnectionMind has been deployed in Meta's large-scale recommendation pipeline and has been evaluated through online A/B tests, achieving a 0.43% improvement in video watch time. These results demonstrate measurable real-world impact in a production recommendation system.

Problem

• social networks
• personalized recommendation
• large language models
• heterogeneous content
• multi-relational context

Innovation

• graph reasoning
• large language models
• social recommendation
• reinforcement learning
• heterogeneous graph