Relation Classification on VERHallu
86.9RC-CausalHuman
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Human2026.01 | 86.9 | 85.2 | 87.6 | |
| Qwen-VL-2.5-72BParameters=72B2026.01 | 31 | 34.1 | 33.4 | |
| Gemini-3-Pro2026.01 | 30 | 35 | 40 | |
| ChatGPT-4o2026.01 | 24.5 | 32 | 33.5 | |
| Qwen-VL-2.5-32BParameters=32B2026.01 | 16.3 | 18 | 29.6 |