Correctness Prediction on GPQA
79.42AccuracyLOCUS
Evaluation Results
| Method | Links | |
|---|---|---|
| LOCUSTrain samples=10242026.01 | 79.42 | |
| LOCUSTrain samples=5122026.01 | 79.18 | |
| LOCUSTrain samples=2562026.01 | 79.17 | |
| EMBEDLLMTrain samples=2562026.01 | 79.03 | |
| EMBEDLLMTrain samples=10242026.01 | 79.02 | |
| IRT-NETTrain samples=10242026.01 | 79.02 | |
| kNNTrain samples=10242026.01 | 78.93 | |
| IRT-NETTrain samples=2562026.01 | 78.87 | |
| KMEANSTrain samples=2562026.01 | 78.76 | |
| EMBEDLLMTrain samples=5122026.01 | 78.73 | |
| EMBEDLLM+Train samples=2562026.01 | 78.69 | |
| KMEANSTrain samples=10242026.01 | 78.67 | |
| IRT-NETTrain samples=5122026.01 | 78.59 | |
| EMBEDLLM+Train samples=10242026.01 | 78.47 | |
| kNNTrain samples=5122026.01 | 78.21 | |
| EMBEDLLM+Train samples=5122026.01 | 78.08 | |
| KMEANSTrain samples=5122026.01 | 76.37 | |
| kNNTrain samples=2562026.01 | 74.43 |