Object In-View Check on TEA (test)
62.39AccuracyHuman
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Human2026.02 | 62.39 | — | |
| Gemini-2.5-Pro2026.02 | 49.57 | — | |
| GPT-52026.02 | 48.72 | — | |
| Gemini-2.5-Flash2026.02 | 45.3 | — | |
| llama-4-Scout2026.02 | 44.44 | — | |
| o3-20252026.02 | 43.59 | — | |
| GPT-4o2026.02 | 42.74 | — | |
| Claude-Opus-42026.02 | 42.74 | — | |
| GPT-4.12026.02 | 18.8 | — |