Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Meta-Reinforcement Learning research benchmarks · SOTA2 Research

Meta-Reinforcement Learning

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Hopper-Param (ID)

302Average Return

30

Jun 2, 2026

Cheetah-Vel-Sparse (OOD)

296Average Return

15

Jun 2, 2026

Walker Param-Sparse

82Average Return

10

Jun 2, 2026

Point-Robot Sparse

42Average Return

10

Jun 2, 2026

140Average Return

10

Jun 2, 2026

60Average Return

10

Jun 2, 2026

236Average Return

10

Jun 2, 2026

Walker-Param (ID)

158Average Return

10

Jun 2, 2026

269Average Return

10

Jun 2, 2026

MuJoCo HalfCheetah 10D-task (c) (test)

1,024CVaR 0.05 Return

7

Feb 26, 2026

MuJoCo HalfCheetah 10D-task (b) (test)

1,697CVaR0.05 Return

7

Feb 26, 2026

MuJoCo HalfCheetah 10D-task (a) (test)

1,227CVaR 0.05 Return

7

Feb 26, 2026

MuJoCo HalfCheetah Body variation (test)

935CVaR 0.05 Return

7

Feb 26, 2026

MuJoCo HalfCheetah Mass variation (test)

1,259CVaR 0.05 Return

7

Feb 26, 2026

MuJoCo HalfCheetah Velocity variation (test)

-184CVaR 0.05 Return

7

Feb 26, 2026

Meta-World in-distribution v2 (test)

0Assembly Success Rate

6

Mar 4, 2026

Point-Robot Sparse (OOD)

10Average Return

5

Jun 2, 2026

Point-Robot-Sparse (ID)

36Average Return

5

Jun 2, 2026

59Average Return

5

Jun 2, 2026

269Average Return

5

Jun 2, 2026

45Success Rate

5

Feb 26, 2026

Meta-World ML1 max 5M steps

6825th Percentile Success Rate

4

Jun 8, 2026

MetaWorld ML-10

48.7Reach

4

Feb 26, 2026

MetaWorld ML-10 (test)

2,144.3Drawer Open

4

Feb 26, 2026

MetaWorld ML-10 (train)

3,696.9Reach

4

Feb 26, 2026

Page 1 of 2

PreviousNext

Popular datasets

MuJoCo HalfCheetah Metaworld cheetah MuJoCo Ant MuJoCo Humanoid

Follow for updates

Join SOTA2 Contact