Abstention in Question Answering on KUQ Cont. Subjective
87.2Abstention F1TIAR
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| TIARModel=Qwen3-8B2026.05 | 87.2 | 90.8 | 83.8 | 95.7 | |
| DPOModel=Qwen3-8B2026.05 | 87.1 | 93.2 | 81.8 | 96.4 | |
| R-TuningModel=Qwen3-8B2026.05 | 87 | 91.7 | 82.8 | 95.7 | |
| TruthRLModel=Qwen3-8B2026.05 | 86.7 | 91.7 | 82.2 | 96.7 | |
| RFTModel=Llama-3.1-8B-Instruct2026.05 | 77.7 | 70.9 | 86 | 67.8 | |
| TIARModel=Llama-3.1-8B-Instruct2026.05 | 76.2 | 69.7 | 83.9 | 70.7 | |
| R-TuningModel=Llama-3.1-8B-Instruct2026.05 | 75 | 67.7 | 84.1 | 66.3 | |
| TruthRLModel=Llama-3.1-8B-Instruct2026.05 | 74.8 | 66.5 | 85.5 | 70.7 | |
| DPOModel=Llama-3.1-8B-Instruct2026.05 | 74.6 | 68.2 | 82.1 | 66.7 | |
| RFTModel=Qwen3-8B2026.05 | 72.4 | 68.2 | 77.2 | 88 |