List your company

Procedural Reasoning on RecipeQA

46.6BertScoreGPT-4o
15.60823.65431.739.746May 14, 2026

Evaluation results

MethodLinks
GPT-4oFramework=CoP2026.0546.649.7
Gemini-2.5Framework=Baseline2026.0542.847.2
GPT-4oFramework=Baseline2026.0538.748.6
Gemini-2.5Framework=CoP2026.0537.844.6
Qwen2.5-72BFramework=CoP2026.0526.532.3
Qwen2.5-72BFramework=Baseline2026.0523.935
GPT-4o-miniFramework=CoP2026.052229.6
GPT-4o-miniFramework=Baseline2026.0520.731.9
Claude-3-5Framework=CoP2026.0519.333.6
Claude-3-5Framework=Baseline2026.0516.826.3