Meta 推出 ConnectionMind：结合社交网络与 LLM 的推荐框架
原标题：ConnectionMind: Leveraging Social Networks and Large Language Models for Personalized Recommendation at Meta
AI 摘要
Meta 提出 ConnectionMind 框架，将社交网络结构与大型语言模型（LLM）结合，用于个性化推荐。该框架构建异构图，将推荐视为图推理问题，采用两阶段学习（监督微调和强化学习）训练推理策略。在 Meta 大规模推荐系统中部署后，在线 A/B 测试显示视频观看时长提升 0.43%。
正文节选
ConnectionMind: Leveraging Social Networks and Large Language Models for Personalized Recommendation at Meta 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 i