| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| 13 Strategic Scenarios (test) | MAFP | TicTacToe Score0.336 | 9 | 1mo ago | |
| Mastermind Extreme | G-Memory | Average Return0.356 | 9 | 4mo ago | |
| Mastermind Hard | ProcMEM | Average Return0.463 | 9 | 4mo ago | |
| FreeCiv Map Seed 2025 (test) | Mastaba | Input Tokens (k)14 | 6 | 25d ago | |
| TextArena SPIN-Bench | MEMO | Win Rate (Simple Negotiation)54.9 | 6 | 4mo ago | |
| Simple Hanabi held-out (test) | MARSHAL | Win Rate37.27 | 2 | 4mo ago | |
| Leduc Hold'em held-out (test) | MARSHAL | Win Rate0.5389 | 2 | 4mo ago | |
| Connect Four held-out (test) | MARSHAL | Win Rate21.55 | 2 | 4mo ago | |
| Mini Hanabi (train) | MARSHAL | Win Rate55.28 | 2 | 4mo ago | |
| Kuhn Poker (train) | MARSHAL | Win Rate44.49 | 2 | 4mo ago | |
| Tic-Tac-Toe (train) | MARSHAL | Win Rate54.05 | 2 | 4mo ago |