Meta Reinforcement Learning on Meta-World in-distribution v2 (test) benchmark leaderboard · SOTA2 Research

Meta Reinforcement Learning on Meta-World in-distribution v2 (test)

0Assembly Success RateSPC

Updated 5mo ago

Evaluation Results

Method	Links
SPC2026.03		0	—	8.3	1.7	26.7	100	83.3	90	96.7	100	3.3	25	88.3	25	100	95	98.3	96.7	100	50	96.7	90	33.3	30	98.3	95	68.3	75	25	41.7	71.7	25	1.7	0	56.7	38.3	74	75	38.3	25	60	6.7	6.7	15	38.3	6.7	0	88.3	88.3	100	100
CSRO2026.03		0	—	0	3.3	11.7	95	68.3	55	83.3	98.3	1.7	18.3	93.3	16.7	100	86.7	93.3	90	100	33.3	95	90	20	16.7	91.7	88.3	45	78.3	21.7	23.3	68.3	26.7	3.3	5	56.7	23.3	66.7	61.7	13.3	20	21.7	5	3.3	5	18.3	0	5	75	68.3	95	88.3
DORA2026.03		0	—	0	0	1.7	90	75	36.7	86.7	96.7	0	13.3	90	8.3	100	93.3	93.3	91.7	98.3	31.7	83.3	68.3	33.3	26.7	93.3	85	26.7	36.7	23.3	25	51.7	18.3	0	1.7	48.3	20	74	61.7	20	25	41.7	3.3	5	6.7	21.7	0	0	91.7	65	95	78.3
FOCAL2026.03		0	—	0	0	6.7	98.3	78.3	35	91.7	95	1.7	13.3	90	5	100	81.7	90	90	98.3	30	83.3	88.3	36.7	25	93.3	93.3	40	68.3	25	18.3	60	30	0	1.7	60	15	68	53.3	16.7	16.7	21.7	1.7	0	5	20	0	1.7	76.7	58.3	98.3	81.7
UNICORN-SS2026.03		0	—	0	0	3.3	91.7	83.3	35	91.7	100	0	18.3	86.7	18.3	100	90	93.3	96.7	98.3	21.7	90	80	40	15	91.7	93.3	36.7	66.7	25	26.7	58.3	28.3	1.7	3.3	53.3	15	63.3	66.7	8.3	21.7	21.7	6.7	4	5	10	0	1.7	56.7	60	100	73.3
UNICORN-SUP2026.03		0	—	0	0	11.7	90	68.3	31.7	81.7	100	3.3	18.3	65	6.7	96.7	85	61.7	86.7	98.3	45	78.3	65	40	25	93.3	88.3	50	58.3	28.3	23.3	50	0	0	0	63.3	8.3	76.7	63.3	13.3	16.7	36.7	1.7	5	1.7	30	1.7	5	88.3	85	98.3	81.7