List your company
ResearchFormerly WizWand

Pocket Cube solving on Pocket Cube (test)

77.41Multi AccuracyXoT (w/ 1 r)
-2.909217.942938.79559.6471Nov 7, 2023
Updated 8mo ago

Evaluation results

MethodLinks
XoT (w/ 1 r)LLM=GPT-4, LLM invoked=4.08, f_theta invoked=122.542023.1177.411.72
XoT (w/ 1 r)LLM=GPT-3.5, LLM invoked=4.13, f_theta invoked=115.732023.1148.722.2
GoT (k=3)LLM=GPT-4, LLM invoked=13.362023.1116.852.77
ToT (b=3)LLM=GPT-4, LLM invoked=56.582023.116.522.99
ToT (b=3)LLM=GPT-3.5, LLM invoked=58.722023.115.832.99
IOLLM=GPT-4, LLM invoked=1.002023.111.091.98
GoT (k=3)LLM=GPT-3.5, LLM invoked=14.762023.111.092.99
CoTLLM=GPT-4, LLM invoked=1.002023.110.821.91
CoT-SCLLM=GPT-4, LLM invoked=1.002023.110.822.92
CoTLLM=GPT-3.5, LLM invoked=1.002023.110.551.05
IOLLM=GPT-3.5, LLM invoked=1.002023.110.272
CoT-SCLLM=GPT-3.5, LLM invoked=10.002023.110.182.9