List your company

Navigation Reasoning on Gridworld o.o.d 20x20 (out-of-domain)

91.5Pass@1TSLM
2.5825.66548.7571.835Jan 30, 2026

Evaluation results

MethodLinks
TSLMevaluation_protocol=pass@12026.0191.5—
Procedure Cloningevaluation_protocol=pass@12026.0181.1—
SCevaluation_protocol=pass@12026.0133—
GRPOevaluation_protocol=pass@12026.016—
Tree-of-Thoughtevaluation_protocol=pass@1002026.01—42.7