List your company

ResearchFormerly WizWand

Reacher

Benchmarks

TaskDatasetSOTA resultTrendResults

Continuous Control

fixedReacher

0.849Average Reward

6

Reacher Hard

86Success Rate

5

Target Reaching

Reacher Yellow (target task)

19.8Success Count

5

Target Reaching

Reacher Green (target task)

19.4Target Reached Count

5

Target Reaching

Reacher Red target task

19.8Target Reach Count

5

Goal-conditioned planning

Reacher

90Success Rate

5

Pixel-based goal-conditioned control

Reacher

87.2Success Rate

5

Reinforcement Learning

Reacher Easy OOD

158.5Score (mass×0.7)

5

Continuous Control

Reacher MuJoCo v5

3Max Evaluation Return

5

Reinforcement Learning

Reacher

6.48Maximum Return

5

Continuous Robotic Control

Reacher normal v2 (test)

-2.39Final Performance

5

Extrapolative Generalization

Reacher-hard Unseen Goal

967.64Mean Reward

5

Extrapolative Generalization

Reacher-hard Unseen View

321.9Mean Reward

5

Reinforcement Learning Surrogate Modeling

Reacher (R) (test)

81Reward Ratio

4

articulated control

Reacher

90Success Rate

4

Target Reaching

Reacher Blue (source task)

20Target Reaches Count

4

Reacher (test)

81.33Planning Success Rate

4

Nominal Planning

Reacher held-out trajectories 57 (test)

83.5Success Rate

4

Goal-conditioned continuous reaching

Reacher

100Success Rate

4

Reinforcement Learning

Reacher v4

-4Avg Episodic Reward

4

Reinforcement Learning

Reacher

-11.69IQM Returns

4

Reinforcement Learning

Reacher target domain 1.5 density

-7.1Reward

4

Reinforcement Learning

Reacher target 0.5 gravity

-7.2Reward

4

Reinforcement Learning

Reacher 1.5 gravity MuJoCo (test)

-8.3Reward

4

Reinforcement Learning

Reacher MuJoCo (broken source)

-7.2Reward

4