List your company
ResearchFormerly WizWand

Logical Reasoning on BOARD (BOARDGAMEQA)

96.3AccuracyGemini 2.5 Pro+ASP
41.90856.02970.1584.271May 16, 2023Nov 12, 2023May 10, 2024Nov 6, 2024May 5, 2025Nov 1, 2025Apr 30, 2026
Updated 6mo ago

Evaluation results

MethodLinks
Gemini 2.5 Pro+ASPInference Pipeline=LLM+ASP, Prompt Reference Type=Compact Reference2026.0496.3
Gemini 2.5 Flash+ASPInference Pipeline=LLM+ASP, Prompt Reference Type=Compact Reference2026.0494.7
Gemini 2.5 FlashInference Pipeline=Baseline2026.0493
DS-R1-0528+ASPInference Pipeline=LLM+ASP, Prompt Reference Type=Compact Reference2026.0491.7
Gemini 2.5 ProInference Pipeline=Baseline2026.0489.8
o4-miniInference Pipeline=Baseline2026.0488.2
o4-mini+ASPInference Pipeline=LLM+ASP, Prompt Reference Type=Compact Reference2026.0485.2
DS-R1-0528Inference Pipeline=Baseline2026.0481
SATLMLanguage Model=code-davinci-002, Decoding Method=self-consistency decoding2023.0580.7
SATLMLanguage Model=code-davinci-002, Decoding Method=greedy decoding2023.0579.4
DS-V3Inference Pipeline=Baseline2026.0471.3
COTLanguage Model=code-davinci-002, Decoding Method=self-consistency decoding2023.0562.8
COTLanguage Model=code-davinci-002, Decoding Method=greedy decoding2023.0560.7
STANDARDLanguage Model=code-davinci-002, Decoding Method=greedy decoding2023.0544.6
DS-V3+ASPInference Pipeline=LLM+ASP, Prompt Reference Type=Compact Reference2026.0444