VP-QA on ProcedureVQA
62.5AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oPrompting Strategy=Zero-Shot2026.05 | 62.5 | |
| Claude-3-5-SonnetPrompting Strategy=CoT2026.05 | 55.6 | |
| GPT-4oPrompting Strategy=CoT2026.05 | 51.8 | |
| Claude-3-5-SonnetPrompting Strategy=Zero-Shot2026.05 | 47.5 | |
| Gemini-2.5-FlashPrompting Strategy=Zero-Shot2026.05 | 44.7 | |
| Qwen2.5-VL-72BPrompting Strategy=Zero-Shot2026.05 | 43.4 | |
| GPT-4o-miniPrompting Strategy=CoT2026.05 | 42.7 | |
| Gemini-2.5-FlashPrompting Strategy=CoT2026.05 | 42.6 | |
| Qwen2.5-VL-72BPrompting Strategy=CoT2026.05 | 41.3 | |
| GPT-4o-miniPrompting Strategy=Zero-Shot2026.05 | 34.4 | |
| Qwen2.5-VL-7BPrompting Strategy=CoT2026.05 | 26.6 | |
| Qwen2.5-VL-7BPrompting Strategy=Zero-Shot2026.05 | 25.1 |