Compositionality Estimation on Random Index H
267.1L(P+)Gemini-2.0
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini-2.02026.04 | 267.1 | 3.83 | 1.81 | |
| DeepSeek-R12026.04 | 174.07 | 4.17 | 26.54 | |
| DeepSeek-V32026.04 | 145.4 | 8.57 | 12.85 | |
| o1-mini2026.04 | 138.27 | 1.27 | 56.69 | |
| QwQ-Plus2026.04 | 138.23 | 10.97 | 0 | |
| Claude-3.52026.04 | 116.37 | 15.73 | 0 | |
| o3-mini2026.04 | 106.9 | 0.5 | 73.2 | |
| Gemini-2.52026.04 | 87.57 | 0.93 | 76.58 | |
| Qwen-Max2026.04 | 51 | 14.83 | 1.38 | |
| GPT-4o2026.04 | 45.33 | 15.8 | 0 | |
| Claude-3.72026.04 | 40.23 | 3.13 | 79.04 |