Multi-agent policy generation on Leduc hold'em repeated
49.3Population ReturnCSRO-AlphaEvolve
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CSRO-AlphaEvolveFramework=CSRO, Oracle=AlphaEvolve2026.03 | 49.3 | 4.4 | 44.9 | |
| CSRO-LinearRefinementFramework=CSRO, Oracle=LinearRefinement2026.03 | 43.8 | 9.8 | 34 | |
| CSRO-ZeroShotFramework=CSRO, Oracle=ZeroShot2026.03 | 40.4 | 19.6 | 20.7 | |
| CFR+2026.03 | 39.8 | 0 | 39.8 | |
| PSRO-IMPALA2026.03 | 13.3 | 58.4 | 45 |