List your company

Planning on Tower of Hanoi (held-out)

72.7AccuracyMemRL
5.122.6540.257.75May 27, 2026

Evaluation results

MethodLinks
MemRLtraining samples=1002026.0572.7
MemRLtraining samples=52026.0551.7
GEPAtraining samples=52026.0543.3
COREtraining samples=1002026.0542.7
COREtraining samples=102026.0542.3
COREtraining samples=52026.0540
MemRLtraining samples=102026.0539.3
GEPAtraining samples=1002026.0535.3
GEPAtraining samples=102026.0531
Episodic RAGtraining samples=1002026.0530.3
Episodic RAGtraining samples=52026.0528.7
Episodic RAGtraining samples=102026.0524.3
No Learningtraining samples=52026.0517.9
No Learningtraining samples=102026.0517.9
No Learningtraining samples=1002026.0517.9
GRPOtraining samples=102026.0512
GRPOtraining samples=1002026.0510.7
GRPOtraining samples=52026.057.7