Hugging Face's logo

Datasets:
EvalData
/
BCCT-Hub

Tasks:
Feature Extraction
Languages:
English
Size:
1K<n<10K
Tags:
representation-similarity
representation-convergence
cross-model-transport
benchmark
alignment
evaluation
License:
Dataset card Files
xet
 Community
BCCT-Hub
File size: 3,126 Bytes
285f7ca
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
{
  "n_pairs": 10,
  "results": [
    {
      "pair": "clip_vitb16 <-> clip_vitl14",
      "tau_atlas": 0.468,
      "perf_linear_r2_perdim": 0.327,
      "perf_linear_cosine": 0.9095,
      "perf_linear_procrustes_dist": 0.5886,
      "cca_mean": 0.6522
    },
    {
      "pair": "resnet50 <-> resnet101",
      "tau_atlas": 0.5445,
      "perf_linear_r2_perdim": 0.0969,
      "perf_linear_cosine": 0.7059,
      "perf_linear_procrustes_dist": 0.5322,
      "cca_mean": 0.9941
    },
    {
      "pair": "dinov2_vitb14 <-> dinov2_vitl14",
      "tau_atlas": 0.4073,
      "perf_linear_r2_perdim": 0.0701,
      "perf_linear_cosine": 0.5592,
      "perf_linear_procrustes_dist": 0.5149,
      "cca_mean": 0.7564
    },
    {
      "pair": "convnext_base <-> convnext_large",
      "tau_atlas": 0.8475,
      "perf_linear_r2_perdim": 0.3942,
      "perf_linear_cosine": 0.7591,
      "perf_linear_procrustes_dist": 0.6265,
      "cca_mean": 0.7126
    },
    {
      "pair": "dinov2_vitb14 <-> clip_vitb16",
      "tau_atlas": 0.0449,
      "perf_linear_r2_perdim": 0.0079,
      "perf_linear_cosine": 0.7826,
      "perf_linear_procrustes_dist": 0.4958,
      "cca_mean": 0.7863
    },
    {
      "pair": "vit_base_patch16_224.mae <-> vit_base_patch16_224",
      "tau_atlas": 0.0024,
      "perf_linear_r2_perdim": 0.0274,
      "perf_linear_cosine": 0.5584,
      "perf_linear_procrustes_dist": 0.3794,
      "cca_mean": 0.4586
    },
    {
      "pair": "beit_base_patch16_224 <-> resnet50",
      "tau_atlas": 0.2277,
      "perf_linear_r2_perdim": 0.0043,
      "perf_linear_cosine": 0.6273,
      "perf_linear_procrustes_dist": 0.4309,
      "cca_mean": 0.7052
    },
    {
      "pair": "clip_vitb32 <-> dinov2_vits14",
      "tau_atlas": 0.0099,
      "perf_linear_r2_perdim": 0.2507,
      "perf_linear_cosine": 0.6024,
      "perf_linear_procrustes_dist": 0.5047,
      "cca_mean": 0.6911
    },
    {
      "pair": "mixer_b16_224 <-> vit_base_patch16_224",
      "tau_atlas": 0.179,
      "perf_linear_r2_perdim": 0.0462,
      "perf_linear_cosine": 0.6604,
      "perf_linear_procrustes_dist": 0.5005,
      "cca_mean": 0.6745
    },
    {
      "pair": "efficientnet_b0 <-> swin_base_patch4_window7_224",
      "tau_atlas": 0.3716,
      "perf_linear_r2_perdim": 0.0579,
      "perf_linear_cosine": 0.6211,
      "perf_linear_procrustes_dist": 0.4959,
      "cca_mean": 0.8218
    }
  ],
  "rank_correlations": {
    "perf_linear_r2_perdim_vs_perf_linear_cosine": {
      "kendall_tau": 0.2,
      "spearman_rho": 0.2727
    },
    "perf_linear_r2_perdim_vs_perf_linear_procrustes_dist": {
      "kendall_tau": 0.7778,
      "spearman_rho": 0.9152
    },
    "perf_linear_r2_perdim_vs_cca_mean": {
      "kendall_tau": 0.0222,
      "spearman_rho": 0.0061
    },
    "perf_linear_cosine_vs_perf_linear_procrustes_dist": {
      "kendall_tau": 0.3333,
      "spearman_rho": 0.4788
    },
    "perf_linear_cosine_vs_cca_mean": {
      "kendall_tau": 0.1111,
      "spearman_rho": 0.1636
    },
    "perf_linear_procrustes_dist_vs_cca_mean": {
      "kendall_tau": 0.1556,
      "spearman_rho": 0.1758
    }
  }
}