Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

puzzle

Benchmarks

Task NameDataset NameSOTA ResultTrend
Offline Reinforcement Learningpuzzle-3x3-play OGBench 5 tasks v0
Average Success Rate87
19
Offline Goal-Conditioned Reinforcement Learningpuzzle 3x3 play v0
Success Rate100
13
overallpuzzle 4x6
Success Rate5,100
10
task5puzzle-4x6
Success Rate0
10
task4puzzle 4x6
Success Rate23
10
task3puzzle-4x6
Success Rate6,700
10
task2puzzle 4x6
Success Rate66
10
task1puzzle 4x6
Success Rate10,000
10
overallpuzzle 4x5
Success Rate9,700
10
task5puzzle 4x5
Success Rate8,800
10
task4puzzle 4x5
Success Rate99
10
task3puzzle 4x5
Success Rate100
10
task2puzzle 4x5
Success Rate9,900
10
task1puzzle 4x5
Success Rate10,000
10
Offline Goal-Conditioned Reinforcement Learningpuzzle-4x6-1B
Success Rate9,100
10
Offline Goal-Conditioned Reinforcement Learningpuzzle 4x5
Success Rate9,600
10
Manipulationpuzzle 4x4-play-oraclerep v0
Task 1 Success Rate62
9
Manipulationpuzzle 3x3 play oraclerep v0
Task 1 Success Rate99
9
Offline Goal-conditioned Reinforcement Learningpuzzle 3x3-play-oraclerep v0
task199
9
Puzzle SolvingPuzzle 10,000 puzzles (test)
Puzzle Success Rate97.2
7
Logic ReasoningPuzzle 1.0 (test)
F1 Score19
7
Charts, figures, and puzzlesPuzzle
Score77.35
6
Offline Goal-Conditioned Reinforcement Learningpuzzle 4x6-play v0
Average Binary Success Rate18
6
Offline Goal-Conditioned Reinforcement Learningpuzzle 4x5 play v0
Success Rate18
6
Offline Goal-Conditioned Reinforcement Learningpuzzle 4x4 play v0
Average Binary Success Rate88
6
Showing 25 of 25 rows