TP on Embodied City
98ASR (%)PPIA
Evaluation results
| Method | Links | |
|---|---|---|
| PPIAModel=gemini-1-p22026.01 | 98 | |
| PPIAModel=gemini-1-fl2026.01 | 98 | |
| PPIAModel=gemini-1-pl2026.01 | 96 | |
| PPIAModel=gpt-4o-mini2026.01 | 92 | |
| PPIAModel=gpt-4o2026.01 | 88 | |
| PPIAModel=claude-3-5-sl2026.01 | 88 | |
| PPIAModel=gpt-4-turbo2026.01 | 72 | |
| PPIAModel=llama3.2-90b-vi2026.01 | 72 | |
| PPIAModel=llama3.2-11b-vision2026.01 | 52 | |
| PPIAModel=claude-3-5-haiku2026.01 | 16 |