List your company

Visual Reasoning on MM-Star (test)

69.1AccuracyVisRef
46.84452.62258.464.178Feb 27, 2026

Evaluation results

MethodLinks
VisRefModel=Qwen3-VL-8B, Configuration=Visual token selection with adaptive stopping2026.0269.1
STModel=Qwen3-VL-8B, Configuration=Standard Thinking (Baseline)2026.0266.5
TSRModel=Qwen3-VL-8B, Configuration=Textual Self-Reflection2026.0265.9
VisRefModel=InternVL3.5-8B, Configuration=Visual token selection with adaptive stopping2026.0263.1
TSRModel=InternVL3.5-8B, Configuration=Textual Self-Reflection2026.0258.3
STModel=InternVL3.5-8B, Configuration=Standard Thinking (Baseline)2026.0257.2
VisRefModel=SAIL-VL2-8B, Configuration=Visual token selection with adaptive stopping2026.0255.3
TSRModel=SAIL-VL2-8B, Configuration=Textual Self-Reflection2026.0248.9
STModel=SAIL-VL2-8B, Configuration=Standard Thinking (Baseline)2026.0247.7