ResearchBenchmarksArithmetic Reasoning on Game of 24 95 (test)Arithmetic Reasoning on Game of 24 95 (test)Follow100Success RateMGRS0.57626.38852.278.012Nov 28, 2025Evaluation resultsMethodMethodLinksSuccess RateMGRSLLM invoked=7.3LLM invoked=7.32025.11100FoT(n=4)LLM invoked=20.6LLM invoked=20.62025.1193.7XoTLLM invoked=1.8LLM invoked=1.82025.1185.4BoTLLM invoked=3.0LLM invoked=3.02025.1182.4ToTLLM invoked=13.7LLM invoked=13.72025.1174AoTLLM invoked=4.0LLM invoked=4.02025.1111.6GoT(k=1)LLM invoked=7.0LLM invoked=7.02025.115.3CoTLLM invoked=1.0LLM invoked=1.02025.114.4CoT-SCLLM invoked=10.0LLM invoked=10.02025.114.4