{"ok":true,"entity":{"slug":"rlhf","entityType":"concept","name":"人間のフィードバックによる強化学習（RLHF）","canonicalName":"RLHF","displayName":"RLHF","category":"AI基礎概念","shortDescription":"人間の評価を報酬として用い、モデルの出力を人間の好みに沿わせる強化学習の手法。","alias":["Reinforcement Learning from Human Feedback","RLHF"],"searchKeywords":["rlhf","人間のフィードバック","強化学習","アライメント"],"website":null,"parentEntity":null,"primaryCluster":"ai-company","secondaryClusters":[],"id":"rlhf","verificationStatus":"draft","updatedAt":"2026-07-10T01:43:25.958Z"},"referenceIndex":["P-01-001","P-02-001","P-04-001"]}