ResearchBenchmarksModel Generation on T2MM Evaluation Dataset (test)Follow69Score 1 Count0-Shot2.4419.723754.28Apr 24, 2026Evaluation ResultsMethodMethodLinksScore 1 CountScore 2 CountScore 3 CountScore 4 CountScore 5 CountScore 6 CountScore 7 CountScore 8 CountNon-compiling Cases CountPerfect Matches CountScore 9+ Count0-ShotPrompting Strategy=0-ShotPrompting Strategy=0-Shot2026.0469849407306922335310N-ShotPrompting Strategy=N-ShotPrompting Strategy=N-Shot2026.04491151626151624123560139T2MMPrompting Strategy=T2MMPrompting Strategy=T2MM2026.045220180190709040