List your company

Large Vision-Language Model Evaluation on Unweighted Average

9.8ECEII
8.17619.13830.141.062May 11, 2026

Evaluation results

MethodLinks
IIBackbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.059.820.865.771.473.770.2
BICRBackbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.0510.820.667.763.874.871.3
BICRBackbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.0516.223.363.366.568.769
PIKBackbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.0517.924.262.671.775.371.7
BICRBackbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0519.422.564.455.659.369.4
PIKBackbone=deepseek-ai/deepseek-vl2, Number of Seeds=52026.0519.423.359.747.949.561.9
SAPLMABackbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.0520.626.263.172.573.869.6
IIBackbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.0522.629.457.268.762.259.9
SAPLMABackbone=deepseek-ai/deepseek-vl2, Number of Seeds=52026.0523.727.259.85348.264.7
PIKBackbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.052529.158.669.568.968.9
SAPLMABackbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.0526.530.560.970.763.764.5
P(True)Backbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0526.625.158.233.85063.2
PIKBackbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0526.727.656.156.558.468.5
CCPSBackbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.053034.255.368.360.956.9
IIBackbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0530.831.747.655.64755.4
PEBackbone=deepseek-ai/deepseek-vl2, Number of Seeds=52026.0531.928.342.950.350.163.1
SAPLMABackbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0532.732.451.757.354.264.1
Self-ProbingBackbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.0533.833.756.36764.170.7
CCPSBackbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0535.335.548.956.758.865.7
PEBackbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.0535.33852.465.856.954.5
CCPSBackbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.0535.748.550.951.764.849.7
PEBackbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0535.733.441.753.946.455.8
PEBackbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.0535.837.852.866.454.553
Self-ProbingBackbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.0536.238.358.567.865.759
P(True)Backbone=OpenGVLab/InternVL3_5-14B-HF, Number of Seeds=52026.0542.441.955.963.971.568.9
P(True)Backbone=deepseek-ai/deepseek-vl2, Number of Seeds=52026.054339.14044.252.465.2
Self-ProbingBackbone=llava-hf/llava-v1.6-vicuna-13b-hf, Number of Seeds=52026.0545.744.647.355.150.254.4
P(True)Backbone=Qwen/Qwen3-VL-8B-Instruct, Number of Seeds=52026.0548.44850.361.370.565.7
Self-ProbingBackbone=deepseek-ai/deepseek-vl2, Number of Seeds=52026.0550.449.839.149.951.352.6