Knowledge & Understanding on IFEval, EQ-Bench, IFBench, SuperGPQA, MMLU-Pro
79CBRC ScoreSuperGPQA
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| SuperGPQAPrompting=Standardized templates, Decoding=Greedy, Evaluation Metric=Exact match or execution-based2026.01 | 79 | 0.05 | 34 | 940 | 0.08 | 43 | 0.07 | 55 | |
| IFEvalPrompting=Standardized templates, Decoding=Greedy, Evaluation Metric=Exact match or execution-based2026.01 | 75 | 0.11 | 23 | 3,787 | 0.1 | 63 | 0.1 | 58 | |
| EQ-BenchPrompting=Standardized templates, Decoding=Greedy, Evaluation Metric=Exact match or execution-based2026.01 | 75 | 0.08 | 27 | 1,782 | 0.17 | 53 | 0.07 | 56 | |
| IFBenchPrompting=Standardized templates, Decoding=Greedy, Evaluation Metric=Exact match or execution-based2026.01 | 71 | 0.11 | 31 | 1,132 | 0.08 | 51 | 0.07 | 55 | |
| MMLU-ProPrompting=Standardized templates, Decoding=Greedy, Evaluation Metric=Exact match or execution-based2026.01 | 65 | 0.1 | 40 | 2,771 | 0.17 | 36 | 0.18 | 51 |