List your company

Rock-Paper-Scissors on Rock-Paper-Scissors (RPS)

3.28RPS@5 ScoreMemory w/ MEMOPILOT
-0.1520.7391.632.521Jun 7, 2026

Evaluation results

MethodLinks
Memory w/ MEMOPILOTPlayer Model=Qwen2.5-14B-Instruct2026.063.28
Memory w/ MEMOPILOTPlayer Model=Qwen3-235B-A22B2026.063.27
Memory w/ DeepSeek-V3.2Player Model=Qwen2.5-14B-Instruct2026.061.64
Memory w/ DeepSeek-V3.2Player Model=Qwen3-235B-A22B2026.061.46
Human-Written Counter-StrategyPlayer Model=Qwen2.5-14B-Instruct2026.061
ReasoningBankPlayer Model=Qwen2.5-14B-Instruct2026.060.81
ReasoningBankPlayer Model=Qwen3-235B-A22B2026.060.81
AWMPlayer Model=Qwen3-235B-A22B2026.060.67
AWMPlayer Model=Qwen2.5-14B-Instruct2026.060.64
ReflexionPlayer Model=Qwen2.5-14B-Instruct2026.060.61
Human-Written Counter-StrategyPlayer Model=Qwen3-235B-A22B2026.060.57
Memory w/ Gemini-3.0-FlashPlayer Model=Qwen3-235B-A22B2026.060.54
ReflexionPlayer Model=Qwen3-235B-A22B2026.060.53
MemoryBankPlayer Model=Qwen3-235B-A22B2026.060.48
Memory w/ Gemini-3.0-FlashPlayer Model=Qwen2.5-14B-Instruct2026.060.45
No MemoryPlayer Model=Qwen3-235B-A22B2026.060.44
No MemoryPlayer Model=Qwen2.5-14B-Instruct2026.060.43
MemoryBankPlayer Model=Qwen2.5-14B-Instruct2026.060.43
Memory w/ Qwen2.5-7B-InstructPlayer Model=Qwen2.5-14B-Instruct2026.060.36
Memory w/ Qwen2.5-14B-InstructPlayer Model=Qwen3-235B-A22B2026.060.34
Memory w/ Qwen2.5-14B-InstructPlayer Model=Qwen2.5-14B-Instruct2026.060.21
Memory w/ Qwen2.5-7B-InstructPlayer Model=Qwen3-235B-A22B2026.060.19
Full HistoryPlayer Model=Qwen3-235B-A22B2026.060.03
ExpeLPlayer Model=Qwen2.5-14B-Instruct2026.060.03
Full HistoryPlayer Model=Qwen2.5-14B-Instruct2026.060.02
ExpeLPlayer Model=Qwen3-235B-A22B2026.06-0.02