Mathematical Reasoning on Math Reasoning Overall
63.8Mean@16IRDS
Evaluation Results
| Method | Links | |
|---|---|---|
| IRDSModel=QWEN3-4B, Data Budget=10%2026.05 | 63.8 | |
| PPL-MIDDLEModel=QWEN3-4B, Data Budget=10%2026.05 | 62.3 | |
| LIMRModel=QWEN3-4B, Data Budget=10%2026.05 | 62.3 | |
| DEPOModel=QWEN3-4B, Data Budget=10%2026.05 | 61.8 | |
| IFDModel=QWEN3-4B, Data Budget=10%2026.05 | 59.7 | |
| PPL-TOPModel=QWEN3-4B, Data Budget=10%2026.05 | 59.3 | |
| RANDOMModel=QWEN3-4B, Data Budget=10%2026.05 | 54.7 | |
| TOKEN-LENGTHModel=QWEN3-4B, Data Budget=10%2026.05 | 51.5 | |
| IRDSModel=QWEN3-1.7B, Data Budget=10%2026.05 | 51.4 | |
| LIMRModel=QWEN3-1.7B, Data Budget=10%2026.05 | 50 | |
| DEPOModel=QWEN3-1.7B, Data Budget=10%2026.05 | 48.6 | |
| PPL-MIDDLEModel=QWEN3-1.7B, Data Budget=10%2026.05 | 45.7 | |
| RANDOMModel=QWEN3-1.7B, Data Budget=10%2026.05 | 45.4 | |
| PPL-TOPModel=QWEN3-1.7B, Data Budget=10%2026.05 | 44.3 | |
| IFDModel=QWEN3-1.7B, Data Budget=10%2026.05 | 44.2 | |
| TOKEN-LENGTHModel=QWEN3-1.7B, Data Budget=10%2026.05 | 41.3 | |
| IRDSModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 21.9 | |
| PPL-MIDDLEModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 21.2 | |
| RANDOMModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 21.1 | |
| LIMRModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 21.1 | |
| IFDModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 21 | |
| DEPOModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 20.5 | |
| TOKEN-LENGTHModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 19.9 | |
| PPL-TOPModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.05 | 19.2 |