ResearchBenchmarksLanguage-conditioned visual reasoning on VLABench (test)Follow76Precision Score (Toy)pi_051.0457.526470.48Nov 26, 2025Evaluation ResultsMethodMethodLinksPrecision Score (Toy)Precision Score (Fruit)Precision Score (Painting)Precision Score (Poker)Precision Score (Mahjong)Average Precision Scorepi_02025.11767216108.1436.43E02025.1176651277.332551.07pi_0 FAST2025.117268264233.3348.27pi_0.52025.1152493628.6717.3945.77