List your company

Multi-hop Question Answering on 2WikiMultiHopQA (RAG/CoT Evaluation)

65.78AccuracyRIRS
24.398435.141745.88556.6283Jan 14, 2025

Evaluation results

MethodLinks
RIRSModel=Qwen-Plus2025.0165.7878.9874.6831.88
RIRSModel=Mixed2025.0162.0676.9369.6536.74
ChameleonModel=Qwen-Plus2025.0160.1268.9272.8631.87
RouterDCModel=Qwen-Plus2025.0152.1262.262.4735.63
RIRSModel=LLaMA-3.1-8B2025.0145.6572.8551.6229.6
ChameleonModel=LLaMA-3.1-8B2025.0140.2767.8145.8928.41
CoTModel=Qwen-Plus2025.0137.31——37.31
RouterDCModel=LLaMA-3.1-8B2025.0130.5968.3334.2522.69
CoTModel=LLaMA-3.1-8B2025.0125.99——25.99