ResearchBenchmarksRecommendation Explanation Evaluation on Human Expert Evaluation Set Aggregated (test)Follow4.38Helpfulness ScoreMATRAG3.06963.40983.754.0902Feb 11, 2026Evaluation ResultsMethodMethodLinksHelpfulness ScoreTrustworthiness ScoreInformativeness ScorePersonalization ScoreAverage Evaluation ScoreMATRAG2026.024.384.314.524.244.36G-CRS2026.023.913.824.083.673.87K-RagRec2026.023.783.673.923.513.72MACRec2026.023.563.423.713.283.49Chat-Rec2026.023.122.983.342.873.08