List your company

Question Answering on CommonsenseQA ConceptNet (20% test)

81.3AccuracySAKE
50.82858.73966.6574.561May 21, 2025

Evaluation results

MethodLinks
SAKEModel=Qwen2.5 7B-Instruct2025.0581.3
GRPOModel=Qwen2.5 7B-Instruct2025.0579.7
SFTModel=Qwen2.5 7B-Instruct2025.0579.3
Search-R1Model=Qwen2.5 7B-Instruct2025.0578.5
RAGModel=Qwen2.5 7B-Instruct2025.0577.6
I/O PromptModel=Qwen2.5 7B-Instruct2025.0576.4
GIVEModel=GPT-3.5 Turbo2025.0574.7
SAKEModel=Qwen2.5 3B-Instruct2025.0573.2
ToGModel=GPT-3.5 Turbo2025.0569.8
SFTModel=Qwen2.5 3B-Instruct2025.0567.5
GRPOModel=Qwen2.5 3B-Instruct2025.0566.7
Search-R1Model=Qwen2.5 3B-Instruct2025.0566.7
CoTModel=Qwen2.5 3B-Instruct2025.0565
I/O PromptModel=Qwen2.5 3B-Instruct2025.0564.2
CoTModel=Qwen2.5 7B-Instruct2025.0563.4
RAGModel=Qwen2.5 3B-Instruct2025.0552