List your company

Visual Grounding on VIEW2SPACE v1

69.34mIoUOurs (Grounded CoT)
-2.773615.948234.6753.3918Mar 17, 2026

Evaluation results

MethodLinks
Ours (Grounded CoT)Category=Fine-tuned, Backbone=Qwen3VL-4B2026.0369.3470.92
Ours (CoT)Category=Fine-tuned, Backbone=Qwen3VL-4B2026.0349.7752.34
Ours (Instruct-tuning)Category=Fine-tuned, Backbone=Qwen3VL-4B2026.0347.5850.07
Qwen3-VL-4BCategory=Open-source MLLMs2026.0316.5318.72
Ours (Grounded CoT, Direct Answer at Test)Category=Fine-tuned, Backbone=Qwen3VL-4B2026.0316.3511.5
Qwen3-VL-2BCategory=Open-source MLLMs2026.0311.25.07
Qwen3-VL-8BCategory=Open-source MLLMs2026.0310.057.37
GPT-5.2Category=Closed-source MLLMs2026.038.742.84
GPT-5Category=Closed-source MLLMs2026.038.183.43
GPT-5-miniCategory=Closed-source MLLMs2026.037.422.77
GPT-4oCategory=Closed-source MLLMs2026.033.60.56
GPT-5-nanoCategory=Closed-source MLLMs2026.032.890.87
DeepSeek-VL2-SmallCategory=Open-source MLLMs2026.032.860.19
Qwen2.5-VL-7BCategory=Open-source MLLMs2026.032.680.32
Gemma-3-12B-itCategory=Open-source MLLMs2026.032.490.24
Random (frequency)Category=Baseline2026.032.010.27
Molmo2-8BCategory=Open-source Spatial Models2026.031.510
Mantis-8B (SigLip)Category=Open-source MLLMs2026.031.240.23
Molmo2-4BCategory=Open-source Spatial Models2026.031.170.13
OpenviewCategory=Open-source Spatial Models2026.031.080.3
Internvl2.5-8BCategory=Open-source MLLMs2026.030.790
Idefics2-8BCategory=Open-source MLLMs2026.030.390
Random (chance)Category=Baseline2026.030.220
MINDCUBE-PlainCategory=Open-source Spatial Models2026.030.120
RoBobrain2.0Category=Open-source Spatial Models2026.0300
Spatial-MLLMCategory=Open-source Spatial Models2026.0300
MINDCUBE-AugCategory=Open-source Spatial Models2026.0300