ResearchBenchmarksOutcome Reasoning on RNN-TopoFollow88.9M' F1 MeanGPT-564.25270.65177.0583.449May 17, 2025Evaluation ResultsMethodMethodLinksM' F1 MeanY' F1 MeanGPT-5Model=GPT-5Model=GPT-52025.0588.983.4GPT-o4Model=GPT-o4Model=GPT-o42025.0587.981.6Llama4-MModel=Llama4-MModel=Llama4-M2025.0580.575DeepSeekModel=DeepSeekModel=DeepSeek2025.0578.372Gemini2.5Model=Gemini2.5Model=Gemini2.52025.0577.170.6Qwen3Model=Qwen3Model=Qwen32025.0575.669.4Llama4-SModel=Llama4-SModel=Llama4-S2025.0565.258.7