Refusal on Irrelevant Queries on ROPES
94.66Refusal F1Doc2Atom
Evaluation Results
| Method | Links | |
|---|---|---|
| Doc2AtomBase Model=Gemma-2-2B-It2026.06 | 94.66 | |
| Doc2AtomBase Model=Qwen3-4B-Instruct2026.06 | 76.94 | |
| D2LckptBase Model=Qwen3-4B-Instruct2026.06 | 50.07 | |
| D2LrawBase Model=Qwen3-4B-Instruct2026.06 | 12.14 | |
| D2LckptBase Model=Gemma-2-2B-It2026.06 | 11.19 | |
| D2LatomBase Model=Gemma-2-2B-It2026.06 | 5.48 | |
| D2LatomBase Model=Qwen3-4B-Instruct2026.06 | 5.48 | |
| D2LrawBase Model=Gemma-2-2B-It2026.06 | 5.24 |