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

Kuhn Poker

Benchmarks

Task NameDataset NameSOTA ResultTrend
Exploitability computation256-Kuhn poker (full)
Exploitability0.007
30
Multi-Agent Strategic ReasoningKuhn Poker In-domain
Success Rate65.6
8
Game PlayingKuhn Poker
Raw Reward0.98
6
Solving Zero-Sum Partially Observable Stochastic Games (zs-POSGs)Kuhn Poker
Time0.01
6
CFR solver performanceKuhn poker
Average Iteration Time1.04
4
Strategic game playingKuhn Poker (train)
Win Rate44.49
2
Showing 6 of 6 rows