Instruction-driven history-aware policies for robotic manipulations
Open paper
Benchmark
Model
Rank
Results
robot-manipulation-generalization-on-gembench
Hiveformer
#7
Average Success Rate: 30.4
Average Success Rate (L1): 60.3±1.5
…
robot-manipulation-on-rlbench
Hiveformer
#13
Succ. Rate (18 tasks, 100 demo/task): 45.3
…
Rank counts only results with a code link.