Behavioral Prediction on Woodward-like Behavioral Prediction Animate
85AccuracyQwen 3.5 Plus
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen 3.5 Plus2025.03 | 85 | |
| GPT-5.22025.03 | 84 | |
| Humans2025.03 | 75 | |
| Gemini 3.1 Pro2025.03 | 63 | |
| Claude 3.5 Sonnet2025.03 | 53 | |
| Qwen VL Max2025.03 | 48 | |
| Gemini 3.1 Flash2025.03 | 41 | |
| Claude Opus 4.62025.03 | 37 | |
| GPT-4o2025.03 | 35 |