Underspecification Detection on ODUTQA-MDC
82.6FROM AccKimi K2
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Kimi K2Clarification Setting=Fixed2026.04 | 82.6 | — | — | 87.95 | 69.02 | 65.54 | 55.51 | |
| Kimi K2Clarification Setting=Dynamic2026.04 | 81.63 | — | — | 87.31 | 64.81 | 59.55 | 52.69 | |
| DoubaoClarification Setting=Fixed2026.04 | 81.36 | — | — | 89.46 | 73.72 | 75.69 | 61.21 | |
| DoubaoClarification Setting=Dynamic2026.04 | 81.22 | — | — | 89.23 | 83.59 | 84.38 | 68.5 | |
| GLM 4Clarification Setting=Fixed2026.04 | 81.1 | — | — | 86.71 | 44.77 | 28.96 | 37.97 | |
| Qwen3 32BClarification Setting=Dynamic2026.04 | 78.78 | — | — | 85.11 | 73.81 | 72.49 | 59.42 | |
| GLM 4Clarification Setting=Dynamic2026.04 | 78.55 | — | — | 84.39 | 57.02 | 53.41 | 46.23 | |
| Qwen3 32BClarification Setting=Fixed2026.04 | 77.66 | — | — | 82.82 | 69.59 | 66.02 | 54.96 | |
| Qwen3 30BClarification Setting=Fixed2026.04 | 75.17 | — | — | 85.1 | 75.67 | 78.99 | 58.55 | |
| Qwen3 30BClarification Setting=Dynamic2026.04 | 69.95 | — | — | 77.45 | 72.49 | 69.5 | 52.92 | |
| BERT2026.04 | — | 99.78 | 99.22 | — | — | — | — |