List your company

Causal Spatial Reasoning on CausalSpatial Level 2

78.57Collision AnticipationHuman
5.738824.646943.55562.4631Jan 19, 2026

Evaluation results

MethodLinks
Human2026.0178.5797.2297.562.583.54—
Gemini2.5 Flash2026.0142.3158.3325.9355.4746.531.79
GPT52026.0140.945062.9664.9654.521.08
GPT5-mini2026.0135.4347.225067.8851.413.94
Qwen3-VL-30B-A3B-T.Scale=30B-A3B, Reasoning_Trained=true2026.0135.0963.8946.348.1844.790.36
Gemini2.5 Pro2026.0130.6552.7848.1564.9648.850.36
Qwen2.5-VL-7B-I.Scale=7B2026.0130.441.6724.0738.6933.795.73
Qwen3-VL-30B-A3B-I.Scale=30B-A3B2026.0130.1652.7831.4852.5541.321.08
Sonnet 3.72026.0129.4148.6134.2660.5844.1617.29
Spatial-MLLM2026.0124.4144.4412.9624.8224.560.72
Qwen3-VL-4B-I.Scale=4B2026.0124.4144.4416.6762.7740.1110.39
Qwen3-VL-8B-T.Scale=8B, Reasoning_Trained=true2026.0123.7544.4422.2234.3129.711.79
Qwen3-VL-8B-I.Scale=8B2026.0122.7636.1129.6350.3635.857.89
Qwen3-VL-2B-I.Scale=2B2026.0119.3538.8922.2237.9628.981.79
Spatial-VLM2026.0115.7541.6724.0737.9628.257.17
LLaVAOneVision1.5-8B-I.Scale=8B2026.0115.3841.6720.3738.6927.8423.3
Random2026.018.5433.3320.3732.1424.5—