List your company

Procedural Reasoning on ProcedureVQA

61.4BertScoreGPT-4o
28.84837.29945.7554.201May 14, 2026

Evaluation results

MethodLinks
GPT-4oFramework=CoP2026.0561.467.2
Qwen2.5-72BFramework=CoP2026.0551.749.6
Claude-3-5Framework=CoP2026.0548.847.8
GPT-4o-miniFramework=CoP2026.0548.348.7
Gemini-2.5Framework=CoP2026.0548.147.3
Gemini-2.5Framework=Baseline2026.0546.443.6
Claude-3-5Framework=Baseline2026.0544.745.6
GPT-4oFramework=Baseline2026.0543.554.2
GPT-4o-miniFramework=Baseline2026.0538.640.2
Qwen2.5-72BFramework=Baseline2026.0530.140.6