Procedural Reasoning on ProcedureVQA
61.4BertScoreGPT-4o
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-4oFramework=CoP2026.05 | 61.4 | 67.2 | |
| Qwen2.5-72BFramework=CoP2026.05 | 51.7 | 49.6 | |
| Claude-3-5Framework=CoP2026.05 | 48.8 | 47.8 | |
| GPT-4o-miniFramework=CoP2026.05 | 48.3 | 48.7 | |
| Gemini-2.5Framework=CoP2026.05 | 48.1 | 47.3 | |
| Gemini-2.5Framework=Baseline2026.05 | 46.4 | 43.6 | |
| Claude-3-5Framework=Baseline2026.05 | 44.7 | 45.6 | |
| GPT-4oFramework=Baseline2026.05 | 43.5 | 54.2 | |
| GPT-4o-miniFramework=Baseline2026.05 | 38.6 | 40.2 | |
| Qwen2.5-72BFramework=Baseline2026.05 | 30.1 | 40.6 |