BlocksWorld
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Blocksworld (test)
98.2Accuracy
35
Blocksworld
97Blocksworld Accuracy
26
BlocksWorld
100Success Rate
20
BlocksWorld
100Success Rate
17
BlocksWorld (BW) (test)
51Loop Ratio
17
Blocksworld
55Scale
12
Blocks (Blocksworld)
100Accuracy
12
Blocksworld unseen problems
100Completion Rate
11
Blocksworld known optimal problems
1Optimal Rate
11
Blocksworld 5-7
30.5Completion Rate
10
Blocksworld
100Completion Rate
9
Blocksworld Misalign variant
0Token Count
7
Blocksworld Randomized variant
100Success Rate
7
Blocksworld Mystery variant
100Success Rate
7
Blocksworld Plain variant
100Success Rate
7
blocksworld advanced
100Coverage
6
blocksworld original
100Coverage
6
Blocksworld 30 tasks Autoscale (test)
16Coverage
6
Blocksworld GTOHP
30Coverage
6
Blocksworld (test)
0.39Average Solving Time (s)
5
Blocksworld n=6
80Success Rate
4
Blocksworld n=5
100Success Rate (SR)
4
Blocksworld n=4
90Success Rate (%)
4
Blocksworld (n=3)
100Success Rate
4
Blocksworld Planning
589.3Ntokens
4