Confidence Estimation on SummEval
0.717AUROCVERDI LR ensemble
Evaluation Results
| Method | Links | |
|---|---|---|
| VERDI LR ensembleCalls=1×2026.05 | 0.717 | |
| Best single signalCalls=1×2026.05 | 0.675 | |
| CoT-UQCalls=5×2026.05 | 0.652 | |
| Trace lengthCalls=1×2026.05 | 0.561 | |
| Logprob confidenceCalls=1×2026.05 | 0.371 |