Stereo_Depth / analysis /dataset_level_analysis /summary_by_model_task.csv
LLDDSS's picture
Upload folder using huggingface_hub
18104f9 verified
Raw History Blame Contribute Delete
2.21 kB
model,task,n,scoring,chance,floor,acc_b0,acc_mean,acc_min,acc_max,acc_range,acc_std,slope_per_baseline,delta_b025_minus_b0,headroom_captured_mean,above_floor
Qwen_Qwen2.5-VL-3B-Instruct,vsr,2195,generate,0.5,0.5,0.7713,0.7595,0.7481,0.7713,0.0232,0.008,-0.0929,-0.0232,0.5191,True
Qwen_Qwen2.5-VL-3B-Instruct,youtube_level1,821,generate,0.25,0.25,0.3995,0.4046,0.3995,0.4117,0.0122,0.005,-0.0077,0.0012,0.2061,True
Qwen_Qwen2.5-VL-3B-Instruct,youtube_level2,764,generate,0.25,0.25,0.3626,0.3743,0.3626,0.3809,0.0183,0.0063,0.0344,0.0079,0.1658,True
Qwen_Qwen2.5-VL-3B-Instruct,youtube_level3,836,generate,0.25,0.25,0.2667,0.2548,0.2392,0.2667,0.0275,0.0102,0.0287,-0.0012,0.0064,True
Qwen_Qwen2.5-VL-7B-Instruct,vsr,2195,generate,0.5,0.5,0.8223,0.805,0.7882,0.8223,0.0342,0.0101,-0.1047,-0.0342,0.61,True
Qwen_Qwen2.5-VL-7B-Instruct,youtube_level1,821,generate,0.25,0.25,0.447,0.4519,0.4421,0.4665,0.0244,0.0077,0.0097,0.0049,0.2692,True
Qwen_Qwen2.5-VL-7B-Instruct,youtube_level2,764,generate,0.25,0.25,0.4031,0.3918,0.377,0.4031,0.0262,0.0109,-0.1017,-0.0262,0.1891,True
Qwen_Qwen2.5-VL-7B-Instruct,youtube_level3,836,generate,0.25,0.25,0.2799,0.2779,0.2679,0.2871,0.0191,0.0061,-0.0273,-0.006,0.0372,True
Qwen_Qwen3.5-4B,vsr,2195,generate,0.5,0.5,0.8214,0.8134,0.8046,0.8214,0.0169,0.0062,-0.0695,-0.0169,0.6269,True
Qwen_Qwen3.5-4B,youtube_level1,821,generate,0.25,0.25,0.5274,0.5296,0.5189,0.5408,0.0219,0.0078,0.0564,0.0073,0.3729,True
Qwen_Qwen3.5-4B,youtube_level2,764,generate,0.25,0.25,0.3914,0.3859,0.3796,0.3914,0.0118,0.0042,-0.0337,-0.0065,0.1812,True
Qwen_Qwen3.5-4B,youtube_level3,836,generate,0.25,0.25,0.2931,0.2957,0.2883,0.3038,0.0156,0.005,-0.0144,0.0,0.0609,True
Qwen_Qwen3.5-4B-Base,vsr,2195,likelihood,0.5,0.5,0.8319,0.8273,0.8164,0.8323,0.0159,0.0057,-0.0594,-0.0155,0.6547,True
Qwen_Qwen3.5-4B-Base,youtube_level1,821,likelihood,0.25,0.25,0.542,0.557,0.542,0.5725,0.0305,0.0107,0.0807,0.0134,0.4094,True
Qwen_Qwen3.5-4B-Base,youtube_level2,764,likelihood,0.25,0.25,0.4424,0.4483,0.4398,0.4542,0.0144,0.0053,-0.0097,-0.0026,0.2644,True
Qwen_Qwen3.5-4B-Base,youtube_level3,836,likelihood,0.25,0.25,0.3026,0.2789,0.2667,0.3026,0.0359,0.0117,-0.0622,-0.0203,0.0385,True