List your company

Post-shift accepted-trajectory risk estimation on RAG tool stress (test)

0.3Risk EstimationToolChain-CRC
-0.1042.6235.358.077Jun 16, 2026

Evaluation results

MethodLinks
ToolChain-CRCCalibration Strategy=Trajectory calibration2026.060.3
Fixed trajectoryCalibration Strategy=Fixed trajectory2026.060.4
Prompt/retrieval-onlyCalibration Strategy=Prompt/retrieval-only2026.065
Final-onlyCalibration Strategy=Final-answer-only2026.0610.4