Post-shift accepted-trajectory risk estimation on Multi-seed tool-chain mean
1.9Post-shift Accepted Trajectory RiskToolChain-CRC
Evaluation results
| Method | Links | |
|---|---|---|
| ToolChain-CRCCalibration Strategy=Trajectory calibration2026.06 | 1.9 | |
| Fixed trajectoryCalibration Strategy=Fixed trajectory2026.06 | 2.4 | |
| Final-onlyCalibration Strategy=Final-answer-only2026.06 | 11 |