List your company

Multimodal Reasoning on Multiple Evaluation Benchmarks Aggregate (test)

105.2Relative Average PerformanceGRACE
89.80893.80497.8101.796May 13, 2026

Evaluation results

MethodLinks
GRACEData selection ratio=15%, Backbone=Qwen3-VL-2B2026.05105.2
GRACEData selection ratio=10%, Backbone=Qwen3-VL-2B2026.05103.2
ICONSData selection ratio=10%, Backbone=Qwen3-VL-2B2026.05102.4
ICONSData selection ratio=15%, Backbone=Qwen3-VL-2B2026.05101.7
CADCData selection ratio=10%, Backbone=Qwen3-VL-2B2026.05101.3
CADCData selection ratio=15%, Backbone=Qwen3-VL-2B2026.05100.7
StepmaxData selection ratio=15%, Backbone=Qwen3-VL-2B2026.05100.6
LongestData selection ratio=10%, Backbone=Qwen3-VL-2B2026.05100.5
GRACEData selection ratio=5%, Backbone=Qwen3-VL-2B2026.05100.2
FullData selection ratio=5%, Backbone=Qwen3-VL-2B2026.05100
FullData selection ratio=10%, Backbone=Qwen3-VL-2B2026.05100
FullData selection ratio=15%, Backbone=Qwen3-VL-2B2026.05100
RandomData selection ratio=15%, Backbone=Qwen3-VL-2B2026.05100
CADCData selection ratio=5%, Backbone=Qwen3-VL-2B2026.0599.8
LESSData selection ratio=15%, Backbone=Qwen3-VL-2B2026.0598.9
StepmaxData selection ratio=10%, Backbone=Qwen3-VL-2B2026.0598.8
RandomData selection ratio=10%, Backbone=Qwen3-VL-2B2026.0598.5
LongestData selection ratio=5%, Backbone=Qwen3-VL-2B2026.0598.3
LongestData selection ratio=15%, Backbone=Qwen3-VL-2B2026.0598.2
RandomData selection ratio=5%, Backbone=Qwen3-VL-2B2026.0596.4
ICONSData selection ratio=5%, Backbone=Qwen3-VL-2B2026.0595.9
StepmaxData selection ratio=5%, Backbone=Qwen3-VL-2B2026.0595.6
LESSData selection ratio=10%, Backbone=Qwen3-VL-2B2026.0595.5
LESSData selection ratio=5%, Backbone=Qwen3-VL-2B2026.0590.4