Share your thoughts, 1 month free Claude Pro on usSee more

Single-trigger control on CMS Run 283408 (test-time fine-tuned)

0.021MAE

GFPO-F

Updated 1mo ago

Evaluation Results

Method	Links
GFPO-F 2026.06		0.021	0.053	67.6	75.64	40.37	76.183	42.91
DQN 2026.06		0.027	0.06	52.7	74.971	39.66	75.743	43.621
PPO 2026.06		0.027	0.061	54	75.06	39.7	75.6	43.15
ADT 2026.06		0.027	0.063	52	75.011	39.51	75.614	42.93
GRPO 2026.06		0.027	0.061	54.1	75.059	39.7	75.601	43.153
L-GRPO 2026.06		0.027	0.061	54.1	75.059	39.68	75.597	43.52
CPO 2026.06		0.027	0.061	54.1	75.059	39.703	75.6	43.15
GFPO-FR 2026.06		0.027	0.06	59.5	75.58	40.24	76.397	44.163
PID 2026.06		0.035	0.08	40.5	75.053	39.191	76.21	44.573
DSPOT 2026.06		0.104	0.18	13.5	67.843	27.225	80.979	56.686
Constant 2026.06		0.142	0.195	0	62.548	14.399	-	-
DQN-F 2026.06		0.142	0.195	0	62.548	14.399	-	-