Scores on benchmarks

Model rank shown below is with respect to all public models.
.257 average_vision rank 157
81 benchmarks
.257
0
ceiling
best
median
.350 neural_vision rank 18
38 benchmarks
.350
0
ceiling
best
median
.491 V1 rank 8
24 benchmarks
.491
0
ceiling
best
median
.689 Coggan2024_fMRI.V1-rdm v1 rank 4
.689
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.550 Marques2020 [reference] rank 325
22 benchmarks
.550
0
ceiling
best
median
.426 V1-orientation rank 334
7 benchmarks
.426
0
ceiling
best
median
.321 Marques2020_Ringach2002-or_selective v1 rank 336
.321
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.215 Marques2020_Ringach2002-circular_variance v1 rank 353
.215
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.233 Marques2020_Ringach2002-orth_pref_ratio v1 rank 331
.233
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.745 Marques2020_Ringach2002-cv_bandwidth_ratio v1 rank 256
.745
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.649 Marques2020_DeValois1982-pref_or v1 rank 336
.649
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.563 Marques2020_Ringach2002-opr_cv_diff v1 rank 325
.563
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.254 Marques2020_Ringach2002-or_bandwidth v1 rank 336
.254
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.763 V1-spatial_frequency rank 197
3 benchmarks
.763
0
ceiling
best
median
.873 Marques2020_DeValois1982-peak_sf v1 rank 20
.873
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.428 Marques2020_Schiller1976-sf_bandwidth v1 [reference] rank 328
.428
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.989 Marques2020_Schiller1976-sf_selective v1 [reference] rank 40
.989
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.378 V1-response_selectivity rank 333
4 benchmarks
.378
0
ceiling
best
median
.280 Marques2020_FreemanZiemba2013-texture_selectivity v1 [reference] rank 331
.280
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.198 Marques2020_Ringach2002-modulation_ratio v1 rank 331
.198
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.868 Marques2020_FreemanZiemba2013-texture_variance_ratio v1 [reference] rank 72
.868
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.164 Marques2020_FreemanZiemba2013-texture_sparseness v1 [reference] rank 336
.164
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.183 V1-receptive_field_size rank 329
2 benchmarks
.183
0
ceiling
best
median
.085 Marques2020_Cavanaugh2002-grating_summation_field v1 [reference] rank 327
.085
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.281 Marques2020_Cavanaugh2002-surround_diameter v1 [reference] rank 307
.281
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.840 V1-surround_modulation rank 22
1 benchmark
.840
0
ceiling
best
median
.840 Marques2020_Cavanaugh2002-surround_suppression_index v1 [reference] rank 22
.840
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.747 V1-texture_modulation rank 62
2 benchmarks
.747
0
ceiling
best
median
.876 Marques2020_FreemanZiemba2013-abs_texture_modulation_index v1 [reference] rank 43
.876
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.618 Marques2020_FreemanZiemba2013-texture_modulation_index v1 [reference] rank 287
.618
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.516 V1-response_magnitude rank 331
3 benchmarks
.516
0
ceiling
best
median
.339 Marques2020_FreemanZiemba2013-max_texture v1 [reference] rank 333
.339
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.817 Marques2020_Ringach2002-max_dc v1 rank 318
.817
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.393 Marques2020_FreemanZiemba2013-max_noise v1 [reference] rank 328
.393
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.233 FreemanZiemba2013.V1-pls v2 [reference] rank 334
.233
0
ceiling
best
median
recordings from 102 sites in V1
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.379 V2 rank 5
2 benchmarks
.379
0
ceiling
best
median
.570 Coggan2024_fMRI.V2-rdm v1 rank 5
.570
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.189 FreemanZiemba2013.V2-pls v2 [reference] rank 404
.189
0
ceiling
best
median
recordings from 103 sites in V2
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.263 V4 rank 351
5 benchmarks
.263
0
ceiling
best
median
.079 Coggan2024_fMRI.V4-rdm v1 rank 39
.079
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.281 SanghaviJozwik2020.V4-pls v1 [reference] rank 351
.281
0
ceiling
best
median
recordings from 50 sites in V4
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.400 Sanghavi2020.V4-pls v1 [reference] rank 354
.400
0
ceiling
best
median
recordings from 47 sites in V4
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.101 SanghaviMurty2020.V4-pls v1 [reference] rank 365
.101
0
ceiling
best
median
recordings from 46 sites in V4
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.456 MajajHong2015.V4-pls v3 [reference] rank 388
.456
0
ceiling
best
median
recordings from 88 sites in V4
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.268 IT rank 232
7 benchmarks
.268
0
ceiling
best
median
.039 Bracci2019.anteriorVTC-rdm v1 rank 189
.039
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.834 Coggan2024_fMRI.IT-rdm v1 rank 8
.834
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.152 SanghaviMurty2020.IT-pls v1 [reference] rank 367
.152
0
ceiling
best
median
recordings from 29 sites in IT
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.282 Sanghavi2020.IT-pls v1 [reference] rank 365
.282
0
ceiling
best
median
recordings from 88 sites in IT
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.320 SanghaviJozwik2020.IT-pls v1 [reference] rank 346
.320
0
ceiling
best
median
recordings from 26 sites in IT
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.251 MajajHong2015.IT-pls v3 [reference] rank 395
.251
0
ceiling
best
median
recordings from 168 sites in IT
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
X Kar2019-ost v2 [reference] rank X
X
0
ceiling
best
median
recordings from 424 sites in IT
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.163 behavior_vision rank 184
43 benchmarks
.163
0
ceiling
best
median
.108 Rajalingham2018-i2n v2 [reference] rank 382
.108
0
ceiling
best
median
match-to-sample task
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.041 Geirhos2021-error_consistency [reference] rank 289
17 benchmarks
.041
0
ceiling
best
median
.022 Geirhos2021colour-error_consistency v1 [reference] rank 278
.022
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.016 Geirhos2021contrast-error_consistency v1 [reference] rank 284
.016
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.050 Geirhos2021cueconflict-error_consistency v1 [reference] rank 280
.050
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.007 Geirhos2021edge-error_consistency v1 [reference] rank 296
.007
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.089 Geirhos2021eidolonI-error_consistency v1 [reference] rank 266
.089
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.098 Geirhos2021eidolonII-error_consistency v1 [reference] rank 259
.098
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.076 Geirhos2021eidolonIII-error_consistency v1 [reference] rank 272
.076
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.034 Geirhos2021falsecolour-error_consistency v1 [reference] rank 254
.034
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.053 Geirhos2021highpass-error_consistency v1 [reference] rank 150
.053
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.024 Geirhos2021lowpass-error_consistency v1 [reference] rank 277
.024
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.037 Geirhos2021phasescrambling-error_consistency v1 [reference] rank 235
.037
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.014 Geirhos2021powerequalisation-error_consistency v1 [reference] rank 274
.014
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.018 Geirhos2021rotation-error_consistency v1 [reference] rank 271
.018
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.139 Geirhos2021silhouette-error_consistency v1 [reference] rank 247
.139
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.009 Geirhos2021sketch-error_consistency v1 [reference] rank 286
.009
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.008 Geirhos2021stylized-error_consistency v1 [reference] rank 291
.008
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.005 Geirhos2021uniformnoise-error_consistency v1 [reference] rank 282
.005
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.000 Baker2022 rank 158
3 benchmarks
.000
0
ceiling
best
median
.000 Baker2022fragmented-accuracy_delta v1 [reference] rank 142
.000
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.000 Baker2022frankenstein-accuracy_delta v1 [reference] rank 142
.000
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.000 Baker2022inverted-accuracy_delta v1 [reference] rank 54
.000
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.426 Maniquet2024 rank 137
2 benchmarks
.426
0
ceiling
best
median
.473 Maniquet2024-confusion_similarity v1 [reference] rank 98
.473
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.379 Maniquet2024-tasks_consistency v1 [reference] rank 171
.379
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.416 Ferguson2024 [reference] rank 144
14 benchmarks
.416
0
ceiling
best
median
.517 Ferguson2024half-value_delta v1 [reference] rank 105
.517
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.855 Ferguson2024gray_hard-value_delta v1 [reference] rank 38
.855
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.532 Ferguson2024lle-value_delta v1 [reference] rank 80
.532
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.009 Ferguson2024juncture-value_delta v1 [reference] rank 189
.009
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.324 Ferguson2024color-value_delta v1 [reference] rank 162
.324
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.185 Ferguson2024round_v-value_delta v1 [reference] rank 182
.185
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.033 Ferguson2024eighth-value_delta v1 [reference] rank 166
.033
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.563 Ferguson2024quarter-value_delta v1 [reference] rank 57
.563
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.425 Ferguson2024convergence-value_delta v1 [reference] rank 98
.425
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.576 Ferguson2024round_f-value_delta v1 [reference] rank 61
.576
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
1.0 Ferguson2024llh-value_delta v1 [reference] rank 1
1.0
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.573 Ferguson2024circle_line-value_delta v1 [reference] rank 46
.573
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.052 Ferguson2024gray_easy-value_delta v1 [reference] rank 191
.052
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.182 Ferguson2024tilted_line-value_delta v1 [reference] rank 193
.182
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.041 Hebart2023-match v1 rank 200
.041
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.110 BMD2024 rank 150
4 benchmarks
.110
0
ceiling
best
median
.104 BMD2024.dotted_1Behavioral-accuracy_distance v1 rank 126
.104
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.103 BMD2024.texture_1Behavioral-accuracy_distance v1 rank 145
.103
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.105 BMD2024.texture_2Behavioral-accuracy_distance v1 rank 144
.105
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.126 BMD2024.dotted_2Behavioral-accuracy_distance v1 rank 107
.126
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.162 Coggan2024_behavior-ConditionWiseAccuracySimilarity v1 rank 127
.162
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.075 engineering_vision rank 270
25 benchmarks
.075
0
ceiling
best
median
.000 ImageNet-C-top1 [reference] rank 243
4 benchmarks
.000
0
ceiling
best
median
.001 ImageNet-C-digital-top1 v2 [reference] rank 207
.001
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.065 Geirhos2021-top1 [reference] rank 253
17 benchmarks
.065
0
ceiling
best
median
.075 Geirhos2021colour-top1 v1 [reference] rank 251
.075
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.062 Geirhos2021contrast-top1 v1 [reference] rank 255
.062
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.076 Geirhos2021cueconflict-top1 v1 [reference] rank 251
.076
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.056 Geirhos2021edge-top1 v1 [reference] rank 268
.056
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.059 Geirhos2021eidolonI-top1 v1 [reference] rank 262
.059
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.062 Geirhos2021eidolonII-top1 v1 [reference] rank 257
.062
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.062 Geirhos2021eidolonIII-top1 v1 [reference] rank 252
.062
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.068 Geirhos2021falsecolour-top1 v1 [reference] rank 252
.068
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.062 Geirhos2021highpass-top1 v1 [reference] rank 252
.062
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.062 Geirhos2021lowpass-top1 v1 [reference] rank 254
.062
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.069 Geirhos2021phasescrambling-top1 v1 [reference] rank 253
.069
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.059 Geirhos2021powerequalisation-top1 v1 [reference] rank 264
.059
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.074 Geirhos2021rotation-top1 v1 [reference] rank 247
.074
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.075 Geirhos2021silhouette-top1 v1 [reference] rank 254
.075
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.056 Geirhos2021sketch-top1 v1 [reference] rank 268
.056
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.070 Geirhos2021stylized-top1 v1 [reference] rank 248
.070
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.062 Geirhos2021uniformnoise-top1 v1 [reference] rank 255
.062
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.307 Hermann2020 [reference] rank 67
2 benchmarks
.307
0
ceiling
best
median
.076 Hermann2020cueconflict-shape_match v1 [reference] rank 252
.076
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9
.538 Hermann2020cueconflict-shape_bias v1 [reference] rank 31
.538
0
ceiling
best
median
sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9

How to use

from brainscore_vision import load_model
model = load_model("alexnet_training_seed_06")
model.start_task(...)
model.start_recording(...)
model.look_at(...)

Benchmarks bibtex

@inproceedings{santurkar2019computer,
    title={Computer Vision with a Single (Robust) Classifier},
    author={Shibani Santurkar and Dimitris Tsipras and Brandon Tran and Andrew Ilyas and Logan Engstrom and Aleksander Madry},
    booktitle={ArXiv preprint arXiv:1906.09453},
    year={2019}
}
        @article {Marques2021.03.01.433495,
	author = {Marques, Tiago and Schrimpf, Martin and DiCarlo, James J.},
	title = {Multi-scale hierarchical neural network models that bridge from single neurons in the primate primary visual cortex to object recognition behavior},
	elocation-id = {2021.03.01.433495},
	year = {2021},
	doi = {10.1101/2021.03.01.433495},
	publisher = {Cold Spring Harbor Laboratory},
	abstract = {Primate visual object recognition relies on the representations in cortical areas at the top of the ventral stream that are computed by a complex, hierarchical network of neural populations. While recent work has created reasonably accurate image-computable hierarchical neural network models of those neural stages, those models do not yet bridge between the properties of individual neurons and the overall emergent behavior of the ventral stream. One reason we cannot yet do this is that individual artificial neurons in multi-stage models have not been shown to be functionally similar to individual biological neurons. Here, we took an important first step by building and evaluating hundreds of hierarchical neural network models in how well their artificial single neurons approximate macaque primary visual cortical (V1) neurons. We found that single neurons in certain models are surprisingly similar to their biological counterparts and that the distributions of single neuron properties, such as those related to orientation and spatial frequency tuning, approximately match those in macaque V1. Critically, we observed that hierarchical models with V1 stages that better match macaque V1 at the single neuron level are also more aligned with human object recognition behavior. Finally, we show that an optimized classical neuroscientific model of V1 is more functionally similar to primate V1 than all of the tested multi-stage models, suggesting room for further model improvements with tangible payoffs in closer alignment to human behavior. These results provide the first multi-stage, multi-scale models that allow our field to ask precisely how the specific properties of individual V1 neurons relate to recognition behavior.HighlightsImage-computable hierarchical neural network models can be naturally extended to create hierarchical {\textquotedblleft}brain models{\textquotedblright} that allow direct comparison with biological neural networks at multiple scales {\textendash} from single neurons, to population of neurons, to behavior.Single neurons in some of these hierarchical brain models are functionally similar to single neurons in macaque primate visual cortex (V1)Some hierarchical brain models have processing stages in which the entire distribution of artificial neuron properties closely matches the biological distributions of those same properties in macaque V1Hierarchical brain models whose V1 processing stages better match the macaque V1 stage also tend to be more aligned with human object recognition behavior at their output stageCompeting Interest StatementThe authors have declared no competing interest.},
	URL = {https://www.biorxiv.org/content/early/2021/08/13/2021.03.01.433495},
	eprint = {https://www.biorxiv.org/content/early/2021/08/13/2021.03.01.433495.full.pdf},
	journal = {bioRxiv}
}
        @article{Schiller1976,
            author = {Schiller, P. H. and Finlay, B. L. and Volman, S. F.},
            doi = {10.1152/jn.1976.39.6.1352},
            issn = {0022-3077},
            journal = {Journal of neurophysiology},
            number = {6},
            pages = {1334--1351},
            pmid = {825624},
            title = {{Quantitative studies of single-cell properties in monkey striate cortex. III. Spatial Frequency}},
            url = {http://www.ncbi.nlm.nih.gov/pubmed/825624},
            volume = {39},
            year = {1976}
            }
        @article{Freeman2013,
            author = {Freeman, Jeremy and Ziemba, Corey M. and Heeger, David J. and Simoncelli, E. P. and Movshon, J. A.},
            doi = {10.1038/nn.3402},
            issn = {10976256},
            journal = {Nature Neuroscience},
            number = {7},
            pages = {974--981},
            pmid = {23685719},
            publisher = {Nature Publishing Group},
            title = {{A functional and perceptual signature of the second visual area in primates}},
            url = {http://dx.doi.org/10.1038/nn.3402},
            volume = {16},
            year = {2013}
            }
        @article{Cavanaugh2002,
            author = {Cavanaugh, James R. and Bair, Wyeth and Movshon, J. A.},
            doi = {10.1152/jn.00692.2001},
            isbn = {0022-3077 (Print) 0022-3077 (Linking)},
            issn = {0022-3077},
            journal = {Journal of Neurophysiology},
            mendeley-groups = {Benchmark effects/Done,Benchmark effects/*Surround Suppression},
            number = {5},
            pages = {2530--2546},
            pmid = {12424292},
            title = {{Nature and Interaction of Signals From the Receptive Field Center and Surround in Macaque V1 Neurons}},
            url = {http://www.physiology.org/doi/10.1152/jn.00692.2001},
            volume = {88},
            year = {2002}
            }
        @Article{Freeman2013,
                author={Freeman, Jeremy
                and Ziemba, Corey M.
                and Heeger, David J.
                and Simoncelli, Eero P.
                and Movshon, J. Anthony},
                title={A functional and perceptual signature of the second visual area in primates},
                journal={Nature Neuroscience},
                year={2013},
                month={Jul},
                day={01},
                volume={16},
                number={7},
                pages={974-981},
                abstract={The authors examined neuronal responses in V1 and V2 to synthetic texture stimuli that replicate higher-order statistical dependencies found in natural images. V2, but not V1, responded differentially to these textures, in both macaque (single neurons) and human (fMRI). Human detection of naturalistic structure in the same images was predicted by V2 responses, suggesting a role for V2 in representing natural image structure.},
                issn={1546-1726},
                doi={10.1038/nn.3402},
                url={https://doi.org/10.1038/nn.3402}
                }
        @misc{Sanghavi_Jozwik_DiCarlo_2021,
  title={SanghaviJozwik2020},
  url={osf.io/fhy36},
  DOI={10.17605/OSF.IO/FHY36},
  publisher={OSF},
  author={Sanghavi, Sachi and Jozwik, Kamila M and DiCarlo, James J},
  year={2021},
  month={Nov}
}
        @misc{Sanghavi_DiCarlo_2021,
  title={Sanghavi2020},
  url={osf.io/chwdk},
  DOI={10.17605/OSF.IO/CHWDK},
  publisher={OSF},
  author={Sanghavi, Sachi and DiCarlo, James J},
  year={2021},
  month={Nov}
}
        @misc{Sanghavi_Murty_DiCarlo_2021,
  title={SanghaviMurty2020},
  url={osf.io/fchme},
  DOI={10.17605/OSF.IO/FCHME},
  publisher={OSF},
  author={Sanghavi, Sachi and Murty, N A R and DiCarlo, James J},
  year={2021},
  month={Nov}
}
        @article {Majaj13402,
            author = {Majaj, Najib J. and Hong, Ha and Solomon, Ethan A. and DiCarlo, James J.},
            title = {Simple Learned Weighted Sums of Inferior Temporal Neuronal Firing Rates Accurately Predict Human Core Object Recognition Performance},
            volume = {35},
            number = {39},
            pages = {13402--13418},
            year = {2015},
            doi = {10.1523/JNEUROSCI.5181-14.2015},
            publisher = {Society for Neuroscience},
            abstract = {To go beyond qualitative models of the biological substrate of object recognition, we ask: can a single ventral stream neuronal linking hypothesis quantitatively account for core object recognition performance over a broad range of tasks? We measured human performance in 64 object recognition tests using thousands of challenging images that explore shape similarity and identity preserving object variation. We then used multielectrode arrays to measure neuronal population responses to those same images in visual areas V4 and inferior temporal (IT) cortex of monkeys and simulated V1 population responses. We tested leading candidate linking hypotheses and control hypotheses, each postulating how ventral stream neuronal responses underlie object recognition behavior. Specifically, for each hypothesis, we computed the predicted performance on the 64 tests and compared it with the measured pattern of human performance. All tested hypotheses based on low- and mid-level visually evoked activity (pixels, V1, and V4) were very poor predictors of the human behavioral pattern. However, simple learned weighted sums of distributed average IT firing rates exactly predicted the behavioral pattern. More elaborate linking hypotheses relying on IT trial-by-trial correlational structure, finer IT temporal codes, or ones that strictly respect the known spatial substructures of IT ({	extquotedblleft}face patches{	extquotedblright}) did not improve predictive power. Although these results do not reject those more elaborate hypotheses, they suggest a simple, sufficient quantitative model: each object recognition task is learned from the spatially distributed mean firing rates (100 ms) of \~{}60,000 IT neurons and is executed as a simple weighted sum of those firing rates.SIGNIFICANCE STATEMENT We sought to go beyond qualitative models of visual object recognition and determine whether a single neuronal linking hypothesis can quantitatively account for core object recognition behavior. To achieve this, we designed a database of images for evaluating object recognition performance. We used multielectrode arrays to characterize hundreds of neurons in the visual ventral stream of nonhuman primates and measured the object recognition performance of \>100 human observers. Remarkably, we found that simple learned weighted sums of firing rates of neurons in monkey inferior temporal (IT) cortex accurately predicted human performance. Although previous work led us to expect that IT would outperform V4, we were surprised by the quantitative precision with which simple IT-based linking hypotheses accounted for human behavior.},
            issn = {0270-6474},
            URL = {https://www.jneurosci.org/content/35/39/13402},
            eprint = {https://www.jneurosci.org/content/35/39/13402.full.pdf},
            journal = {Journal of Neuroscience}}
        @Article{Kar2019,
                                                    author={Kar, Kohitij
                                                    and Kubilius, Jonas
                                                    and Schmidt, Kailyn
                                                    and Issa, Elias B.
                                                    and DiCarlo, James J.},
                                                    title={Evidence that recurrent circuits are critical to the ventral stream's execution of core object recognition behavior},
                                                    journal={Nature Neuroscience},
                                                    year={2019},
                                                    month={Jun},
                                                    day={01},
                                                    volume={22},
                                                    number={6},
                                                    pages={974-983},
                                                    abstract={Non-recurrent deep convolutional neural networks (CNNs) are currently the best at modeling core object recognition, a behavior that is supported by the densely recurrent primate ventral stream, culminating in the inferior temporal (IT) cortex. If recurrence is critical to this behavior, then primates should outperform feedforward-only deep CNNs for images that require additional recurrent processing beyond the feedforward IT response. Here we first used behavioral methods to discover hundreds of these `challenge' images. Second, using large-scale electrophysiology, we observed that behaviorally sufficient object identity solutions emerged {	extasciitilde}30{	hinspace}ms later in the IT cortex for challenge images compared with primate performance-matched `control' images. Third, these behaviorally critical late-phase IT response patterns were poorly predicted by feedforward deep CNN activations. Notably, very-deep CNNs and shallower recurrent CNNs better predicted these late IT responses, suggesting that there is a functional equivalence between additional nonlinear transformations and recurrence. Beyond arguing that recurrent circuits are critical for rapid object identification, our results provide strong constraints for future recurrent model development.},
                                                    issn={1546-1726},
                                                    doi={10.1038/s41593-019-0392-5},
                                                    url={https://doi.org/10.1038/s41593-019-0392-5}
                                                    }
        @article {Rajalingham240614,
                author = {Rajalingham, Rishi and Issa, Elias B. and Bashivan, Pouya and Kar, Kohitij and Schmidt, Kailyn and DiCarlo, James J.},
                title = {Large-scale, high-resolution comparison of the core visual object recognition behavior of humans, monkeys, and state-of-the-art deep artificial neural networks},
                elocation-id = {240614},
                year = {2018},
                doi = {10.1101/240614},
                publisher = {Cold Spring Harbor Laboratory},
                abstract = {Primates{	extemdash}including humans{	extemdash}can typically recognize objects in visual images at a glance even in the face of naturally occurring identity-preserving image transformations (e.g. changes in viewpoint). A primary neuroscience goal is to uncover neuron-level mechanistic models that quantitatively explain this behavior by predicting primate performance for each and every image. Here, we applied this stringent behavioral prediction test to the leading mechanistic models of primate vision (specifically, deep, convolutional, artificial neural networks; ANNs) by directly comparing their behavioral signatures against those of humans and rhesus macaque monkeys. Using high-throughput data collection systems for human and monkey psychophysics, we collected over one million behavioral trials for 2400 images over 276 binary object discrimination tasks. Consistent with previous work, we observed that state-of-the-art deep, feed-forward convolutional ANNs trained for visual categorization (termed DCNNIC models) accurately predicted primate patterns of object-level confusion. However, when we examined behavioral performance for individual images within each object discrimination task, we found that all tested DCNNIC models were significantly non-predictive of primate performance, and that this prediction failure was not accounted for by simple image attributes, nor rescued by simple model modifications. These results show that current DCNNIC models cannot account for the image-level behavioral patterns of primates, and that new ANN models are needed to more precisely capture the neural mechanisms underlying primate object vision. To this end, large-scale, high-resolution primate behavioral benchmarks{	extemdash}such as those obtained here{	extemdash}could serve as direct guides for discovering such models.SIGNIFICANCE STATEMENT Recently, specific feed-forward deep convolutional artificial neural networks (ANNs) models have dramatically advanced our quantitative understanding of the neural mechanisms underlying primate core object recognition. In this work, we tested the limits of those ANNs by systematically comparing the behavioral responses of these models with the behavioral responses of humans and monkeys, at the resolution of individual images. Using these high-resolution metrics, we found that all tested ANN models significantly diverged from primate behavior. Going forward, these high-resolution, large-scale primate behavioral benchmarks could serve as direct guides for discovering better ANN models of the primate visual system.},
                URL = {https://www.biorxiv.org/content/early/2018/02/12/240614},
                eprint = {https://www.biorxiv.org/content/early/2018/02/12/240614.full.pdf},
                journal = {bioRxiv}
            }
        @article{geirhos2021partial,
              title={Partial success in closing the gap between human and machine vision},
              author={Geirhos, Robert and Narayanappa, Kantharaju and Mitzkus, Benjamin and Thieringer, Tizian and Bethge, Matthias and Wichmann, Felix A and Brendel, Wieland},
              journal={Advances in Neural Information Processing Systems},
              volume={34},
              year={2021},
              url={https://openreview.net/forum?id=QkljT4mrfs}
        }
        @article{BAKER2022104913,
                title = {Deep learning models fail to capture the configural nature of human shape perception},
                journal = {iScience},
                volume = {25},
                number = {9},
                pages = {104913},
                year = {2022},
                issn = {2589-0042},
                doi = {https://doi.org/10.1016/j.isci.2022.104913},
                url = {https://www.sciencedirect.com/science/article/pii/S2589004222011853},
                author = {Nicholas Baker and James H. Elder},
                keywords = {Biological sciences, Neuroscience, Sensory neuroscience},
                abstract = {Summary
                A hallmark of human object perception is sensitivity to the holistic configuration of the local shape features of an object. Deep convolutional neural networks (DCNNs) are currently the dominant models for object recognition processing in the visual cortex, but do they capture this configural sensitivity? To answer this question, we employed a dataset of animal silhouettes and created a variant of this dataset that disrupts the configuration of each object while preserving local features. While human performance was impacted by this manipulation, DCNN performance was not, indicating insensitivity to object configuration. Modifications to training and architecture to make networks more brain-like did not lead to configural processing, and none of the networks were able to accurately predict trial-by-trial human object judgements. We speculate that to match human configural sensitivity, networks must be trained to solve a broader range of object tasks beyond category recognition.}
        }
        @article {Maniquet2024.04.02.587669,
	author = {Maniquet, Tim and de Beeck, Hans Op and Costantino, Andrea Ivan},
	title = {Recurrent issues with deep neural network models of visual recognition},
	elocation-id = {2024.04.02.587669},
	year = {2024},
	doi = {10.1101/2024.04.02.587669},
	publisher = {Cold Spring Harbor Laboratory},
	URL = {https://www.biorxiv.org/content/early/2024/04/10/2024.04.02.587669},
	eprint = {https://www.biorxiv.org/content/early/2024/04/10/2024.04.02.587669.full.pdf},
	journal = {bioRxiv}
}
        @misc{ferguson_ngo_lee_dicarlo_schrimpf_2024,
         title={How Well is Visual Search Asymmetry predicted by a Binary-Choice, Rapid, Accuracy-based Visual-search, Oddball-detection (BRAVO) task?},
         url={osf.io/5ba3n},
         DOI={10.17605/OSF.IO/5BA3N},
         publisher={OSF},
         author={Ferguson, Michael E, Jr and Ngo, Jerry and Lee, Michael and DiCarlo, James and Schrimpf, Martin},
         year={2024},
         month={Jun}
}
        @ARTICLE{Hendrycks2019-di,
   title         = "Benchmarking Neural Network Robustness to Common Corruptions
                    and Perturbations",
   author        = "Hendrycks, Dan and Dietterich, Thomas",
   abstract      = "In this paper we establish rigorous benchmarks for image
                    classifier robustness. Our first benchmark, ImageNet-C,
                    standardizes and expands the corruption robustness topic,
                    while showing which classifiers are preferable in
                    safety-critical applications. Then we propose a new dataset
                    called ImageNet-P which enables researchers to benchmark a
                    classifier's robustness to common perturbations. Unlike
                    recent robustness research, this benchmark evaluates
                    performance on common corruptions and perturbations not
                    worst-case adversarial perturbations. We find that there are
                    negligible changes in relative corruption robustness from
                    AlexNet classifiers to ResNet classifiers. Afterward we
                    discover ways to enhance corruption and perturbation
                    robustness. We even find that a bypassed adversarial defense
                    provides substantial common perturbation robustness.
                    Together our benchmarks may aid future work toward networks
                    that robustly generalize.",
   month         =  mar,
   year          =  2019,
   archivePrefix = "arXiv",
   primaryClass  = "cs.LG",
   eprint        = "1903.12261",
   url           = "https://arxiv.org/abs/1903.12261"
}
        @article{hermann2020origins,
              title={The origins and prevalence of texture bias in convolutional neural networks},
              author={Hermann, Katherine and Chen, Ting and Kornblith, Simon},
              journal={Advances in Neural Information Processing Systems},
              volume={33},
              pages={19000--19015},
              year={2020},
              url={https://proceedings.neurips.cc/paper/2020/hash/db5f9f42a7157abe65bb145000b5871a-Abstract.html}
        }
        

Layer Commitment

Region Layer
V1 classifier.0
V2 classifier.0
V4 features.3
IT classifier.3

Visual Angle

None degrees