List your company

Open Question Answering on OpenQA (Accuracy)

88.2AccuracyTrajectory
60.74467.8727582.128Mar 1, 2026

Evaluation results

MethodLinks
TrajectoryTrain Dataset=OpenQA, Model=Qwen2.5-14b2026.0388.2
TrajectoryTrain Dataset=ARC-E, Model=Qwen2.5-14b2026.0384.4
Linear ProbeTrain Dataset=OpenQA, Model=Qwen2.5-14b2026.0380.31
Linear ProbeTrain Dataset=ARC-E, Model=Qwen2.5-14b2026.0377.83
TrajectoryTrain Dataset=ARC-C, Model=Qwen2.5-14b2026.0377.2
Linear ProbeTrain Dataset=ARC-C, Model=Qwen2.5-14b2026.0375.48
Few-shot AccuracyModel=Qwen2.5-14b2026.0362
Zero-shot AccuracyModel=Qwen2.5-14b2026.0361.8