ResearchBenchmarksThrow Away Paper on Real-world (test)Throw Away Paper on Real-world (test)Follow90Success Rate3PoinTr-3.620.74569.3Mar 9, 2026Evaluation resultsMethodMethodLinksSuccess Rate3PoinTrRollouts=10Rollouts=102026.0390DP3Rollouts=10Rollouts=102026.0310ATMRollouts=10Rollouts=102026.030