Datasets:

MrPatoVera
/

lexical-consensus-data

sample_id float64 0 99	quadrant large_stringclasses 4 values	a large_stringlengths 3 13	b large_stringlengths 3 12	contrast1 large_stringlengths 3 13	contrast2 large_stringlengths 3 13	acc_pair float64 0.64 0.98	acc_pair_cvcl float64 0.41 0.82	d_perc float64 0.14 0.89	d_perc_cvcl float64 0.04 0.63	d_sem float64 0.07 0.9	d_sem_orig float64 0.13 0.89	d_sem_paper float64 0.13 0.89	d_things float64 0.06 0.92 ⌀	d_things_scout float64 0.06 0.92 ⌀	d_things_strict float64 0.06 0.92 ⌀	hsj_n int64 0 217	d_hsj_raw float64 0.49 1 ⌀	d_hsj float64 0.49 1 ⌀	d_hsj_ge1 float64 0.49 1 ⌀	d_hsj_ge30 float64 0.49 0.99 ⌀	hsj_n_origmap int64 0 217	d_hsj_raw_origmap float64 0.49 1 ⌀	d_hsj_origmap float64 0.49 1 ⌀	encoder large_stringclasses 1 value	learner large_stringclasses 1 value	encoder_cvcl large_stringclasses 1 value
50	Q3	apple	cup	crocodile	keyboard	0.9262	0.7269	0.716807	0.340738	0.578947	0.578947	0.578947	0.548875	0.548875	0.548875	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
25	Q2	apple	tractor	mountain	squirrel	0.8551	0.6793	0.799385	0.626777	0.619048	0.619048	0.619048	0.818622	0.818622	0.818622	2	1	null	1	null	2	1	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
51	Q3	aquarium_fish	elephant	cup	willow_tree	0.8367	0.654	0.792498	0.130259	0.25	0.25	0.25	0.192118	0.192118	null	141	0.836879	0.836879	0.836879	0.836879	185	0.854054	0.854054	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
75	Q4	baby	television	lion	chair	0.9029	0.6958	0.671261	0.466539	0.636364	0.619048	0.619048	0.842468	0.842468	0.842468	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
26	Q2	bear	bottle	dolphin	mushroom	0.8838	0.63	0.736696	0.376055	0.636364	0.636364	0.636364	0.908117	0.908117	0.908117	100	0.92	0.92	0.92	0.92	100	0.92	0.92	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
76	Q4	beaver	cloud	bus	elephant	0.8544	0.728	0.62022	0.273625	0.809524	0.764706	0.764706	0.636266	0.636266	0.636266	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
0	Q1	beaver	ray	cloud	possum	0.8289	0.6004	0.515666	0.164203	0.333333	0.578947	0.578947	null	null	0.137114	21	0.619048	0.619048	0.619048	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
77	Q4	beaver	snail	chimpanzee	television	0.94	0.7269	0.474777	0.044225	0.416667	0.809524	0.809524	0.145677	0.145677	0.145677	18	0.722222	0.722222	0.722222	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
1	Q1	bed	chair	cockroach	poppy	0.9709	0.8249	0.614322	0.200193	0.2	0.2	0.2	0.234862	0.234862	0.234862	100	0.49	0.49	0.49	0.49	100	0.49	0.49	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
52	Q3	bed	man	tiger	leopard	0.7589	0.6098	0.748537	0.229606	0.578947	0.578947	0.578947	0.586261	0.586261	0.586261	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
27	Q2	bed	sea	butterfly	crab	0.8522	0.62	0.775448	0.349579	0.733333	0.733333	0.733333	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
2	Q1	bee	beetle	otter	tulip	0.8933	0.548	0.398431	0.080495	0.130435	0.130435	0.130435	0.079821	0.079821	0.079821	193	0.580311	0.580311	0.580311	0.580311	193	0.580311	0.580311	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
53	Q3	bee	elephant	tiger	oak_tree	0.8182	0.6931	0.727096	0.131213	0.44	0.44	0.44	0.147427	0.147427	0.147427	9	0.888889	null	0.888889	null	9	0.888889	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
78	Q4	bee	flatfish	dinosaur	train	0.8353	0.5958	0.619765	0.092181	0.481481	0.8	0.8	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
54	Q3	bee	skyscraper	flatfish	lawn_mower	0.8358	0.5467	0.76619	0.338909	0.6	0.6	0.6	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
28	Q2	bee	table	shark	clock	0.8678	0.6956	0.73877	0.263168	0.619048	0.888889	0.888889	0.91603	0.91603	0.91603	5	0.6	null	0.6	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
3	Q1	beetle	camel	willow_tree	tank	0.8173	0.5829	0.689702	0.184541	0.44	0.44	0.44	0.186543	0.186543	0.186543	21	0.952381	0.952381	0.952381	null	21	0.952381	0.952381	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
79	Q4	beetle	motorcycle	raccoon	crab	0.8151	0.4656	0.692655	0.17516	0.636364	0.636364	0.636364	0.809164	0.809164	0.809164	23	0.956522	0.956522	0.956522	null	23	0.956522	0.956522	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
29	Q2	bottle	cloud	squirrel	pine_tree	0.7967	0.6196	0.795229	0.420506	0.764706	0.75	0.75	0.650403	0.650403	0.650403	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
30	Q2	bottle	crocodile	lion	dolphin	0.852	0.6656	0.751279	0.461892	0.636364	0.636364	0.636364	null	null	null	28	0.964286	0.964286	0.964286	null	28	0.964286	0.964286	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
55	Q3	bottle	keyboard	tiger	snail	0.8662	0.6422	0.821888	0.375603	0.333333	0.333333	0.333333	0.650094	0.650094	null	78	0.74359	0.74359	0.74359	0.74359	78	0.74359	0.74359	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
4	Q1	bottle	television	snail	sea	0.9084	0.8178	0.659404	0.158728	0.4	0.368421	0.368421	0.736652	0.736652	0.736652	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
80	Q4	bowl	porcupine	mountain	beetle	0.8878	0.5876	0.61536	0.356501	0.636364	0.636364	0.636364	0.894874	0.894874	0.894874	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
5	Q1	bowl	spider	caterpillar	porcupine	0.8096	0.5202	0.677936	0.291313	0.6	0.6	0.6	0.896068	0.896068	0.896068	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
56	Q3	boy	poppy	tank	porcupine	0.8296	0.7013	0.81003	0.174069	0.428571	0.428571	0.428571	0.818797	0.818797	0.818797	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
57	Q3	bridge	snake	cattle	cup	0.8569	0.5749	0.790226	0.403729	0.578947	0.578947	0.578947	null	null	null	217	0.870968	0.870968	0.870968	0.870968	217	0.870968	0.870968	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
81	Q4	butterfly	palm_tree	worm	bear	0.8824	0.6256	0.668812	0.147434	0.478261	0.789474	0.789474	0.302257	0.302257	0.302257	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
82	Q4	castle	mountain	can	bicycle	0.9773	0.7036	0.537413	0.253739	0.6	0.625	0.625	null	null	null	54	0.759259	0.759259	0.759259	0.759259	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
83	Q4	castle	squirrel	table	sweet_pepper	0.9547	0.654	0.603679	0.239015	0.636364	0.652174	0.652174	null	null	null	5	1	null	1	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
6	Q1	castle	turtle	kangaroo	hamster	0.8371	0.5896	0.682951	0.331509	0.636364	0.5	0.5	null	null	null	22	0.954545	0.954545	0.954545	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
84	Q4	chair	dinosaur	lion	telephone	0.8767	0.6516	0.693337	0.367643	0.652174	0.652174	0.652174	null	null	null	15	0.933333	0.933333	0.933333	null	15	0.933333	0.933333	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
31	Q2	chair	maple_tree	forest	cockroach	0.8371	0.6713	0.773469	0.594118	0.636364	0.777778	0.777778	0.78495	0.78495	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
7	Q1	chimpanzee	crab	snake	telephone	0.8862	0.6933	0.645166	0.11858	0.5	0.5	0.5	null	null	null	12	0.833333	0.833333	0.833333	null	12	0.833333	0.833333	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
8	Q1	chimpanzee	kangaroo	butterfly	flatfish	0.8353	0.5089	0.654223	0.089529	0.310345	0.310345	0.310345	null	null	null	18	0.666667	0.666667	0.666667	null	18	0.666667	0.666667	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
32	Q2	chimpanzee	skyscraper	orchid	otter	0.8158	0.5431	0.762193	0.328417	0.666667	0.666667	0.666667	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
58	Q3	chimpanzee	spider	otter	mouse	0.7347	0.412	0.733792	0.14484	0.481481	0.481481	0.481481	null	null	null	25	0.96	0.96	0.96	null	25	0.96	0.96	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
33	Q2	chimpanzee	wardrobe	oak_tree	flatfish	0.8109	0.6269	0.807476	0.558315	0.68	0.68	0.68	null	null	null	3	1	null	1	null	3	1	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
34	Q2	clock	forest	ray	pine_tree	0.7958	0.612	0.806727	0.499249	0.882353	0.882353	0.882353	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
85	Q4	clock	lizard	tiger	train	0.8596	0.6896	0.70125	0.368495	0.666667	0.666667	0.666667	0.884099	0.884099	0.884099	71	0.985915	0.985915	0.985915	0.985915	71	0.985915	0.985915	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
9	Q1	clock	motorcycle	streetcar	lobster	0.858	0.698	0.710248	0.285318	0.454545	0.454545	0.454545	0.477401	0.477401	0.477401	8	0.875	null	0.875	null	8	0.875	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
59	Q3	clock	tractor	pickup_truck	butterfly	0.826	0.682	0.800138	0.477549	0.428571	0.428571	0.428571	0.554797	0.554797	0.554797	12	0.833333	0.833333	0.833333	null	12	0.833333	0.833333	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
35	Q2	cloud	rose	raccoon	skunk	0.8462	0.5222	0.759335	0.322083	0.789474	0.777778	0.777778	0.492047	0.492047	0.492047	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
60	Q3	cockroach	fox	cup	bottle	0.8556	0.598	0.784664	0.288962	0.461538	0.461538	0.461538	0.188052	0.188052	0.188052	8	0.875	null	0.875	null	8	0.875	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
61	Q3	couch	tiger	lizard	crab	0.7989	0.4836	0.757521	0.37218	0.68	0.555556	0.555556	0.873286	0.873286	0.873286	10	1	1	1	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
36	Q2	crab	maple_tree	cup	pear	0.8687	0.63	0.721649	0.146236	0.5	0.8	0.8	0.605052	0.605052	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
37	Q2	crab	pickup_truck	pine_tree	whale	0.8369	0.65	0.745085	0.395929	0.666667	0.652174	0.652174	0.897582	0.897582	null	8	1	null	1	null	86	0.965116	0.965116	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
86	Q4	crab	train	trout	keyboard	0.8633	0.5627	0.683321	0.249177	0.619048	0.619048	0.619048	0.89334	0.89334	0.89334	6	1	null	1	null	6	1	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
38	Q2	crocodile	cup	beaver	porcupine	0.7133	0.47	0.747225	0.307417	0.619048	0.619048	0.619048	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
87	Q4	crocodile	flatfish	beaver	tank	0.802	0.5516	0.389993	0.092719	0.357143	0.809524	0.809524	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
10	Q1	crocodile	kangaroo	caterpillar	fox	0.8362	0.4556	0.433903	0.064781	0.307692	0.307692	0.307692	null	null	null	8	1	null	1	null	8	1	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
88	Q4	crocodile	palm_tree	apple	bowl	0.9213	0.6607	0.633182	0.138193	0.5	0.8	0.8	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
11	Q1	crocodile	skunk	man	cup	0.9558	0.6453	0.446824	0.037458	0.333333	0.478261	0.478261	null	null	null	14	0.857143	0.857143	0.857143	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
89	Q4	crocodile	trout	fox	train	0.9347	0.6262	0.489642	0.098244	0.307692	0.8	0.8	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
12	Q1	cup	plate	bridge	seal	0.9529	0.6827	0.544535	0.402825	0.2	0.2	0.2	0.156147	0.156147	0.156147	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
62	Q3	cup	whale	pickup_truck	telephone	0.8491	0.7096	0.785329	0.225942	0.636364	0.529412	0.529412	0.858765	0.858765	0.858765	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
39	Q2	cup	wolf	snail	couch	0.8276	0.5769	0.806575	0.270243	0.636364	0.636364	0.636364	0.879772	0.879772	0.879772	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
63	Q3	dolphin	man	caterpillar	sweet_pepper	0.862	0.5682	0.758323	0.230442	0.52	0.52	0.52	0.673118	0.673118	0.673118	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
90	Q4	flatfish	poppy	mountain	lobster	0.8444	0.632	0.70141	0.137774	0.555556	0.8	0.8	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
91	Q4	forest	skunk	aquarium_fish	bowl	0.9344	0.6804	0.552535	0.141926	0.9	0.846154	0.846154	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
40	Q2	forest	telephone	boy	otter	0.8058	0.5536	0.835355	0.511798	0.866667	0.866667	0.866667	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
64	Q3	house	keyboard	boy	aquarium_fish	0.8751	0.6304	0.755149	0.332505	0.411765	0.411765	0.411765	null	null	null	9	0.777778	null	0.777778	null	9	0.777778	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
13	Q1	house	pickup_truck	dinosaur	shark	0.9569	0.736	0.463699	0.082721	0.5	0.473684	0.473684	null	null	null	14	1	1	1	null	149	0.812081	0.812081	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
65	Q3	keyboard	skunk	train	seal	0.7991	0.5987	0.879711	0.382064	0.652174	0.578947	0.578947	0.848327	0.848327	null	2	1	null	1	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
66	Q3	keyboard	snail	plate	orange	0.8413	0.7051	0.835303	0.338984	0.6	0.6	0.6	0.901294	0.901294	null	3	1	null	1	null	3	1	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
67	Q3	keyboard	worm	apple	woman	0.8402	0.6962	0.814776	0.311003	0.555556	0.555556	0.555556	0.909897	0.909897	null	20	0.7	0.7	0.7	null	20	0.7	0.7	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
14	Q1	lamp	shrew	camel	trout	0.8153	0.5044	0.701458	0.39197	0.636364	0.6	0.6	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
92	Q4	leopard	turtle	skunk	woman	0.8762	0.5613	0.617302	0.136286	0.357143	0.8	0.8	0.074769	0.074769	0.074769	19	0.894737	0.894737	0.894737	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
15	Q1	lion	rabbit	elephant	sweet_pepper	0.8944	0.6589	0.553107	0.064125	0.241379	0.241379	0.241379	0.064375	0.064375	0.064375	34	0.529412	0.529412	0.529412	0.529412	34	0.529412	0.529412	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
93	Q4	lion	raccoon	pickup_truck	boy	0.9564	0.7409	0.504489	0.057987	0.172414	0.84	0.84	0.066139	0.066139	0.066139	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
68	Q3	lion	shark	dolphin	crab	0.7673	0.5216	0.777423	0.298382	0.37931	0.37931	0.37931	0.208811	0.208811	0.208811	10	0.8	0.8	0.8	null	10	0.8	0.8	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
41	Q2	lion	turtle	beetle	squirrel	0.7838	0.4867	0.716742	0.132019	0.357143	0.68	0.68	0.08734	0.08734	0.08734	30	0.9	0.9	0.9	0.9	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
16	Q1	lizard	snail	skunk	aquarium_fish	0.8996	0.4827	0.515372	0.045149	0.416667	0.416667	0.416667	0.175476	0.175476	0.175476	177	0.706215	0.706215	0.706215	0.706215	177	0.706215	0.706215	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
17	Q1	man	squirrel	palm_tree	wolf	0.8398	0.5316	0.582671	0.132905	0.454545	0.454545	0.454545	0.673255	0.673255	0.673255	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
18	Q1	man	turtle	seal	couch	0.7944	0.5878	0.690216	0.161536	0.454545	0.578947	0.578947	0.642393	0.642393	0.642393	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
42	Q2	mouse	palm_tree	keyboard	lion	0.8484	0.6469	0.774515	0.100423	0.5	0.8	0.8	0.72781	0.72781	0.72781	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
19	Q1	mouse	shark	wardrobe	television	0.91	0.7191	0.630985	0.190228	0.333333	0.333333	0.333333	0.163328	0.163328	0.163328	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
20	Q1	mouse	shrew	forest	pine_tree	0.8607	0.5767	0.141426	0.0551	0.153846	0.5	0.5	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
43	Q2	mouse	streetcar	possum	oak_tree	0.7576	0.6491	0.745843	0.346155	0.652174	0.652174	0.652174	0.89288	null	0.89288	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
94	Q4	mushroom	ray	seal	whale	0.7847	0.5151	0.691125	0.244797	0.5	0.888889	0.888889	null	null	0.529481	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
44	Q2	oak_tree	orange	raccoon	plain	0.8276	0.7411	0.821873	0.535269	0.652174	0.647059	0.647059	0.568453	0.568453	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
45	Q2	orange	porcupine	dinosaur	lawn_mower	0.8322	0.5142	0.885331	0.508861	0.68	0.68	0.68	0.765816	0.765816	0.765816	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
69	Q3	orchid	pine_tree	tiger	whale	0.8496	0.6584	0.828322	0.21798	0.36	0.36	0.36	0.2026	0.2026	0.2026	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
46	Q2	orchid	sea	fox	pine_tree	0.8211	0.5929	0.861732	0.459859	0.764706	0.764706	0.764706	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
70	Q3	orchid	seal	keyboard	leopard	0.8649	0.6949	0.767675	0.109452	0.538462	0.538462	0.538462	0.706243	0.706243	0.706243	4	0.75	null	0.75	null	4	0.75	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
71	Q3	orchid	shark	lobster	crab	0.6431	0.4676	0.803978	0.230761	0.538462	0.538462	0.538462	0.691894	0.691894	0.691894	16	1	1	1	null	16	1	1	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
72	Q3	orchid	whale	table	telephone	0.8433	0.6536	0.836087	0.330556	0.538462	0.428571	0.428571	0.671287	0.671287	0.671287	9	1	null	1	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
95	Q4	otter	rocket	dinosaur	couch	0.8873	0.5742	0.611723	0.267262	0.652174	0.789474	0.789474	0.833584	0.833584	0.833584	2	1	null	1	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
96	Q4	otter	skunk	shrew	camel	0.8202	0.5456	0.382485	0.052208	0.071429	0.764706	0.764706	0.071667	0.071667	0.071667	122	0.860656	0.860656	0.860656	0.860656	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
97	Q4	palm_tree	woman	mountain	television	0.8538	0.7916	0.65551	0.1811	0.4	0.692308	0.692308	0.74634	0.74634	0.74634	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
73	Q3	pear	road	man	couch	0.8242	0.6347	0.790498	0.354721	0.555556	0.555556	0.555556	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
21	Q1	plain	rocket	turtle	lobster	0.9196	0.6329	0.500003	0.22423	0.571429	0.571429	0.571429	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
22	Q1	plain	worm	oak_tree	sunflower	0.8871	0.7116	0.706222	0.394081	0.571429	0.571429	0.571429	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
74	Q3	plate	road	lizard	turtle	0.7876	0.4947	0.824232	0.516862	0.411765	0.411765	0.411765	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
98	Q4	rabbit	sea	cockroach	bridge	0.8209	0.6671	0.697814	0.346958	0.789474	0.789474	0.789474	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
47	Q2	ray	sweet_pepper	tulip	forest	0.8402	0.6349	0.764475	0.30685	0.833333	0.888889	0.888889	null	null	0.755137	10	0.8	0.8	0.8	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
48	Q2	rose	sea	worm	house	0.8229	0.6922	0.79536	0.498032	0.75	0.75	0.75	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
23	Q1	shrew	squirrel	woman	tank	0.9762	0.7273	0.238103	0.064866	0.153846	0.5	0.5	null	null	null	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
24	Q1	spider	woman	possum	squirrel	0.7553	0.4598	0.675995	0.190091	0.4	0.4	0.4	0.691421	0.691421	0.691421	0	null	null	null	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
49	Q2	squirrel	telephone	seal	lizard	0.7447	0.4393	0.764119	0.501299	0.636364	0.636364	0.636364	0.895523	null	0.895523	6	1	null	1	null	6	1	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)
99	Q4	table	tractor	boy	cattle	0.8404	0.5822	0.70997	0.127567	0.368421	0.875	0.875	0.739487	0.739487	0.739487	10	0.8	0.8	0.8	null	0	null	null	DINOv2-small (CLS, 384-d, L2)	prototype (normalized centroid), 3-way, 5 shots	CVCL vision (ResNeXt-50, 512-d, L2)

Visual similarity, human similarity and artificial word learning (CIFAR)

Companion dataset for Visual Similarity in Adult Word Extension and Artificial Word Learning (Vera, 2026). Author: Patricio M. Vera (ORCID 0009-0004-5384-5952), Simbionte.org. Code: github.com/patriciomvera/lexical-consensus. Preprint: arXiv link coming soon.

Summary

Tabular companion to analyses that relate a frozen self-supervised visual encoder (DINOv2-small) to human similarity and word-extension behaviour, and to an artificial prototype learner over the same space. This repository holds the CC BY 4.0 configs. A separate repository holds the CIFAR-10H-derived tables under CC BY-NC-SA 4.0.

Config	Unit	Rows	What it is
`pairs_c100`	CIFAR-100 class pair	4,950	DINOv2 / CVCL / WordNet / THINGS / HSJ distances
`learner_c100`	Sampled two-category word	100	Prototype-learner naming accuracy + distances
`xtmem_aggregates`	Image × teaching condition	96	Aggregated adult extension (Lewis & Frank 2018)
`word_norms`	Word	95	Glasgow AoA + our DINOv2 dispersion

Auxiliary files (not HF configs): maps/map_things_v1.json, maps/map_hsj_v1.json, and recipe/ (rebuild instructions for XTMEM aggregates).

How it was built (recipe)

Encoder. DINOv2-small CLS embeddings (384-d), L2-normalised. Default Hugging Face image processor: RGB, bicubic resize of the shorter side to 256, 224×224 centre crop, ImageNet normalisation. Pair distance = cosine distance between class centroids (all CIFAR-100 training images per class).

Learner. Prototype (normalised centroid) in 3-way, 5-shot episodes; naming accuracy on held-out images. The two native contrast classes (contrast1, contrast2) are sampled with the same deterministic seed schedule used in the DINOv2 dissociation experiment and reused for the CVCL re-run; they are episode metadata for both encoders, not CVCL-only.

Human similarity.

THINGS SPoSE 66-d distances via a versioned CIFAR→THINGS map (map_things_v1.json).
ImageNet-HSJ: d_hsj is defined only when hsj_n ≥ 10; raw means and counts are also shipped. Trial-level HSJ data are not shipped (rebuild from OSF cn2s3).
Glasgow Norms age of acquisition (aoa_glasgow). Word-frequency (zipf from wordfreq) is not redistributed (author request not to repackage).
dispersion in word_norms is our own derived value: within-class coherence of DINOv2 embeddings for that CIFAR label (not a Glasgow Norms field).

Adult word extension (XTMEM / Lewis & Frank 2018). Only aggregates are shipped: one row per image × harmonized teaching condition over target-domain trials (in_target = 1). Seven raw condition spellings are unified to four canonical labels (one, three_subordinate, three_basic, three_superordinate), matching the paper's exploratory figure. That yields 96 units (= paper \XUnits).

Suppression rule (formal). Any aggregate cell with n_participants < 5 is omitted from xtmem_aggregates. In the full Lewis & Frank sample no cell falls below this threshold at the published grain.

Why n_participants is 520 in every cell. Each person-session runs 12 trials = 4 teaching conditions × 3 categories. Every test image belongs to one category domain, so with in_target = 1 every session contributes exactly one observation to every image × condition cell. Unique people = 520 (paper HFourPersons); unique sessions = 600 (56 people repeated across experiments). Column n_sessions makes this transparent. The constant is a property of the crossed design, not a broadcast of a global total.

Per-participant / per-trial tables (h4_long.csv, h4_first_session.csv) are not redistributed — see recipe/ to rebuild locally from OSF yekhj.

CVCL. Distances and accuracies from the CVCL vision encoder (MIT-licensed model; weights not included). Same episode design as DINOv2 for the 100 pairs.

What we tried that did not ship

Redistributing per-participant XTMEM derivatives — rejected to keep the paper, the code repository and this dataset consistent, and because stimulus-image provenance is undocumented in the upstream GitHub tree.
Packing wordfreq zipf values — rejected at the author's request.
Shipping CIFAR / ImageNet / THINGS / XTMEM images or embedding caches — out of scope; large and not ours to rehost.
A single shared license covering CIFAR-10H — impossible without forcing NC-SA onto CC BY material; CIFAR-10H tables live in the sister repository.
Publishing hsj_trials.csv.gz (~511k rows) — optional; omitted (rebuild from OSF cn2s3).

Limitations

THINGS / HSJ coverage is partial (see covered_* columns). Class mappings involve judgements; scout / strict / origmap variants are included where they exist.
XTMEM aggregates do not support re-analysis of person-level random effects; use the recipe and the original OSF deposit for that.
No images. Learner results are for frozen encoders, CIFAR, two-category words and shallow learners. Post hoc paper analyses are exploratory.
Research use; not a production model-evaluation suite.

Sources and licenses

Source	License	What enters this repo
Lewis & Frank 2018 (OSF yekhj)	CC BY 4.0	Aggregates only
THINGS (OSF f5rn6)	CC BY 4.0	Distances + map
ImageNet-HSJ (OSF cn2s3)	CC BY 4.0	Distances + map
Glasgow Norms	CC BY 4.0	`aoa_glasgow`
CVCL model	MIT	Distances / accuracies only
Our distances, learner outputs, and `dispersion`	CC BY 4.0 (this dataset)	All remaining columns

Third-party values remain under their own licenses. Attribution in any redistribution must preserve the chain above.

Citation

@misc{vera2026visualsimilarity,
  title        = {Visual Similarity in Adult Word Extension and Artificial Word Learning},
  author       = {Vera, Patricio M.},
  year         = {2026},
  note         = {Affiliation: Simbionte.org. ORCID: 0009-0004-5384-5952.
                  Code: https://github.com/patriciomvera/lexical-consensus}
}

Files

pairs_c100/pairs_c100.parquet
learner_c100/learner_c100.parquet
xtmem_aggregates/xtmem_aggregates.parquet
xtmem_aggregates/condition_label_map.json
word_norms/word_norms.parquet
maps/map_things_v1.json
maps/map_hsj_v1.json
recipe/

Downloads last month: 1