List your company

Multiple-Choice QA on CLOTH

84.8AccuracyDirect-Joint
40.069651.682363.29574.9077Jan 23, 2026

Evaluation results

MethodLinks
Direct-JointBackbone Model=Qwen-2.5-7B, Evaluation Protocol=0-shot2026.0184.8
GTCABackbone Model=Qwen-2.5-7B, Evaluation Protocol=0-shot2026.0183.98
LoRA-onlyBackbone Model=Qwen-2.5-7B, Evaluation Protocol=0-shot2026.0183.71
GTCABackbone Model=Llama-3-8B, Evaluation Protocol=0-shot2026.0182.74
LoRA-onlyBackbone Model=Llama-3-8B, Evaluation Protocol=0-shot2026.0181.38
Direct-JointBackbone Model=Llama-3-8B, Evaluation Protocol=0-shot2026.0180.46
BackboneBackbone Model=Qwen-2.5-7B, Evaluation Protocol=0-shot2026.0174.6
BackboneBackbone Model=Llama-3-8B, Evaluation Protocol=0-shot2026.0141.79