Question Answering on MMLU (Test Error Probability)
0.141Test Error ProbabilityGreedy MI
Evaluation Results
| Method | Links | |
|---|---|---|
| Greedy MIk=6, temperatures=3, runs=2, folds=52026.02 | 0.141 | |
| Greedy MIk=5, temperatures=3, runs=2, folds=52026.02 | 0.142 | |
| Greedy MIk=7, temperatures=3, runs=2, folds=52026.02 | 0.143 | |
| Greedy MIk=4, temperatures=3, runs=2, folds=52026.02 | 0.144 | |
| Greedy MIk=3, temperatures=3, runs=2, folds=52026.02 | 0.148 | |
| Greedy MIk=8, temperatures=3, runs=2, folds=52026.02 | 0.148 | |
| Top-k (Acc)k=5, temperatures=3, runs=2, folds=52026.02 | 0.149 | |
| Term 1 (Rel)k=5, temperatures=3, runs=2, folds=52026.02 | 0.149 | |
| Terms 1+2 (mRMR)k=5, temperatures=3, runs=2, folds=52026.02 | 0.149 | |
| Top-k (Acc)k=6, temperatures=3, runs=2, folds=52026.02 | 0.149 | |
| Term 1 (Rel)k=6, temperatures=3, runs=2, folds=52026.02 | 0.149 | |
| Terms 1+2 (mRMR)k=6, temperatures=3, runs=2, folds=52026.02 | 0.149 | |
| Term 1 (Rel)k=7, temperatures=3, runs=2, folds=52026.02 | 0.149 | |
| Top-k (Acc)k=4, temperatures=3, runs=2, folds=52026.02 | 0.15 | |
| Term 1 (Rel)k=4, temperatures=3, runs=2, folds=52026.02 | 0.15 | |
| Terms 1+2 (mRMR)k=4, temperatures=3, runs=2, folds=52026.02 | 0.15 | |
| Top-k (Acc)k=7, temperatures=3, runs=2, folds=52026.02 | 0.15 | |
| Terms 1+2 (mRMR)k=8, temperatures=3, runs=2, folds=52026.02 | 0.15 | |
| Top-k (Acc)k=1, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Term 1 (Rel)k=1, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Terms 1+2 (mRMR)k=1, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Greedy MIk=1, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Terms 1+2 (mRMR)k=2, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Greedy MIk=2, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Terms 1+2 (mRMR)k=3, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Terms 1+2 (mRMR)k=7, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Top-k (Acc)k=8, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Term 1 (Rel)k=8, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Terms 1+2 (mRMR)k=9, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Greedy MIk=9, temperatures=3, runs=2, folds=52026.02 | 0.151 | |
| Top-k (Acc)k=2, temperatures=3, runs=2, folds=52026.02 | 0.152 | |
| Term 1 (Rel)k=2, temperatures=3, runs=2, folds=52026.02 | 0.152 | |
| Top-k (Acc)k=3, temperatures=3, runs=2, folds=52026.02 | 0.152 | |
| Term 1 (Rel)k=3, temperatures=3, runs=2, folds=52026.02 | 0.152 | |
| Term 1 (Rel)k=9, temperatures=3, runs=2, folds=52026.02 | 0.152 | |
| Top-k (Acc)k=9, temperatures=3, runs=2, folds=52026.02 | 0.153 | |
| Terms 1+2 (mRMR)k=10, temperatures=3, runs=2, folds=52026.02 | 0.155 | |
| Greedy MIk=10, temperatures=3, runs=2, folds=52026.02 | 0.155 | |
| Top-k (Acc)k=10, temperatures=3, runs=2, folds=52026.02 | 0.156 | |
| Term 1 (Rel)k=10, temperatures=3, runs=2, folds=52026.02 | 0.156 | |
| Top-k (Acc)k=11, temperatures=3, runs=2, folds=52026.02 | 0.158 | |
| Term 1 (Rel)k=11, temperatures=3, runs=2, folds=52026.02 | 0.158 | |
| Greedy MIk=11, temperatures=3, runs=2, folds=52026.02 | 0.158 | |
| Terms 1+2 (mRMR)k=11, temperatures=3, runs=2, folds=52026.02 | 0.159 | |
| Top-k (Acc)k=12, temperatures=3, runs=2, folds=52026.02 | 0.164 | |
| Term 1 (Rel)k=12, temperatures=3, runs=2, folds=52026.02 | 0.164 | |
| Greedy MIk=12, temperatures=3, runs=2, folds=52026.02 | 0.164 | |
| Terms 1+2 (mRMR)k=12, temperatures=3, runs=2, folds=52026.02 | 0.166 | |
| Top-k (Acc)k=13, temperatures=3, runs=2, folds=52026.02 | 0.171 | |
| Term 1 (Rel)k=13, temperatures=3, runs=2, folds=52026.02 | 0.171 | |
| Terms 1+2 (mRMR)k=13, temperatures=3, runs=2, folds=52026.02 | 0.171 | |
| Greedy MIk=13, temperatures=3, runs=2, folds=52026.02 | 0.171 |