List your company

Visual Reasoning Robustness on VisArgs specific masking 1.0 (test)

58.78Answer Flip RateGemini-1.5-Flash
57.702464.976272.2579.5238Dec 3, 2025

Evaluation results

MethodLinks
Gemini-1.5-FlashMasking Type=Specific Masking2025.1258.7879.0830.6
GPT-4oMasking Type=Specific Masking2025.1274.7492.8635.51
Llama-3.2-90B-Vision-InstructMasking Type=Specific Masking2025.1275.7293.7349.97
Qwen-2.5-VL-7B-InstructMasking Type=Specific Masking2025.1285.7295.5926.57