Identification on Personal Workspace
70.11CoverageTIDE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TIDEBackbone=Gemini 3.5 Flash2026.06 | 70.11 | 71.05 | |
| TIDEBackbone=GPT-5 mini2026.06 | 67.64 | 68.76 | |
| TIDEBackbone=Qwen 3.6 Flash2026.06 | 50.5 | 58.13 | |
| SINGLE-AGENTBackbone=GPT-5 mini2026.06 | 47.85 | 54.63 | |
| SINGLE-AGENTBackbone=Gemini 3.5 Flash2026.06 | 42.99 | 52.95 | |
| MULTI-AGENTBackbone=Gemini 3.5 Flash2026.06 | 38.98 | 49.95 | |
| TIDEBackbone=Claude Sonnet 4.52026.06 | 35.77 | 62.44 | |
| SINGLE-AGENTBackbone=Qwen 3.6 Flash2026.06 | 32.44 | 44.67 | |
| MULTI-AGENTBackbone=Qwen 3.6 Flash2026.06 | 31.04 | 42.27 | |
| MULTI-AGENTBackbone=GPT-5 mini2026.06 | 27.24 | 38.85 | |
| MULTI-AGENTBackbone=Claude Sonnet 4.52026.06 | 18.04 | 36.04 | |
| SINGLE-AGENTBackbone=Claude Sonnet 4.52026.06 | 13.53 | 25.74 |