Visual Question Answering On Gqa Test2019

평가 지표

Accuracy

Binary

Consistency

Distribution

Open

Plausibility

Validity

평가 결과

이 벤치마크에서 각 모델의 성능 결과

비교 표

모델 이름	Accuracy	Binary	Consistency	Distribution	Open	Plausibility	Validity
모델 1	60.98	77.32	90.77	5.36	46.55	84.93	96.38
모델 2	60.59	78.44	92.66	7.28	44.83	85.38	96.57
모델 3	46.55	63.26	74.57	7.46	31.8	84.25	96.02
모델 4	64.92	82.63	94.37	5.11	49.29	84.91	96.64
모델 5	59.37	77.53	88.63	6.06	43.35	84.71	96.18
모델 6	59.8	76.74	89.14	5.11	44.85	84.2	96.23
모델 7	60.17	77.19	89.61	5.83	45.14	84.46	96.36
모델 8	59.29	77.31	88.94	5.8	43.38	84.43	96.3
모델 9	37.03	56.61	63.96	28.4	19.74	85.12	95.76
모델 10	57.35	75.07	87.61	5.94	41.71	84.5	95.86
vinvl-l-enriching-visual-representation-with	64.85	82.59	94.0	4.59	49.19	84.91	96.62
모델 12	55.0	72.09	83.47	5.29	39.92	84.66	96.34
bottom-up-and-top-down-attention-for-image	49.74	66.64	78.71	5.98	34.83	84.57	96.18
모델 14	57.89	74.54	85.45	5.73	43.19	84.99	96.4
graph-reasoning-networks-for-visual-question	61.22	78.69	90.31	6.77	45.81	85.43	96.36
모델 16	56.95	74.62	87.71	5.81	41.36	84.57	95.98
모델 17	60.18	76.97	89.65	5.29	45.36	84.47	96.33
모델 18	63.17	78.94	93.25	3.71	49.25	84.28	96.41
모델 19	55.65	72.86	89.18	9.69	40.46	85.27	96.33
모델 20	56.96	74.97	85.12	7.13	41.06	84.85	96.38
모델 21	60.27	76.99	90.16	5.39	45.51	84.49	96.31
모델 22	60.02	76.37	90.05	5.63	45.59	84.34	96.29
모델 23	55.93	71.81	83.2	6.05	41.93	85.09	96.01
모델 24	59.93	79.09	93.72	10.1	43.02	85.92	96.41
모델 25	63.2	77.91	89.84	5.25	50.22	85.15	96.47
모델 26	89.3	91.2	98.4	0.0	87.4	97.2	98.9
모델 27	59.43	77.11	89.05	6.39	43.82	84.94	96.56
모델 28	56.18	72.84	85.46	5.42	41.47	84.04	96.18
모델 29	44.06	57.57	38.18	8.35	32.13	75.19	85.94
모델 30	59.54	77.98	89.21	6.01	43.26	84.94	96.24
모델 31	61.1	77.99	91.08	5.52	46.19	84.82	96.36
모델 32	56.65	73.65	84.35	6.07	41.64	84.37	95.94
모델 33	56.59	73.0	84.7	4.68	42.11	84.86	96.4
모델 34	51.22	69.36	82.44	6.45	35.2	83.82	96.12
모델 35	60.14	77.15	89.58	5.81	45.12	84.47	96.36
모델 36	57.1	76.0	91.7	10.52	40.41	85.58	96.16
모델 37	61.09	77.84	88.92	5.68	46.3	85.49	96.43
모델 38	60.67	78.02	89.81	6.41	45.36	84.84	96.31
모델 39	53.57	70.15	81.14	5.32	38.94	84.67	96.36
모델 40	61.05	78.02	89.77	5.24	46.06	84.95	96.5
모델 41	62.44	80.28	94.36	5.33	46.69	84.91	96.46
vinvl-making-visual-representations-matter-in	64.65	82.63	94.35	4.72	48.77	84.98	96.62
모델 43	57.21	74.46	87.6	5.6	41.99	84.87	96.2
모델 44	17.82	36.05	62.4	19.99	1.74	34.84	35.78
모델 45	54.15	69.3	82.36	5.41	40.79	85.15	95.99
모델 46	58.06	76.6	90.96	7.6	41.7	85.27	96.31
lxmert-learning-cross-modality-encoder	62.71	79.79	93.1	6.42	47.64	85.21	96.36
모델 48	59.81	78.02	91.43	6.0	43.75	84.77	96.5
모델 49	28.9	42.94	51.69	93.08	16.62	74.81	88.86
모델 50	55.7	72.88	83.52	5.32	40.53	84.81	96.39
모델 51	57.65	75.22	87.35	5.48	42.14	84.73	96.18
모델 52	40.3	61.18	74.11	40.44	21.88	86.13	96.14
모델 53	70.23	77.5	86.94	1.49	63.82	83.77	96.65
모델 54	57.14	75.07	87.36	5.29	41.31	84.49	95.87
모델 55	48.97	63.85	83.85	13.72	35.83	83.93	95.62
모델 56	59.12	76.69	88.9	5.6	43.6	84.78	96.43
모델 57	56.16	73.56	84.99	5.87	40.8	84.83	96.4
모델 58	53.31	70.41	80.33	6.4	38.23	84.32	95.99
모델 59	48.44	65.02	81.19	17.79	33.81	85.29	96.15
모델 60	60.07	76.84	89.32	6.21	45.27	84.55	96.35
모델 61	36.75	55.24	69.93	40.84	20.44	84.13	95.1
모델 62	60.42	77.12	89.69	6.03	45.68	84.56	96.35
모델 63	63.94	80.84	91.54	4.69	49.03	84.74	96.56
모델 64	51.87	67.99	80.2	6.77	37.64	84.35	96.25
모델 65	60.37	77.09	89.77	6.43	45.61	84.56	96.22
모델 66	52.19	69.15	78.34	5.69	37.22	83.44	95.45
모델 67	72.14	81.16	90.96	2.39	64.19	84.81	96.77
모델 68	61.12	78.07	91.13	5.55	46.16	84.8	96.36
모델 69	60.93	77.83	90.3	5.74	46.01	84.69	96.35
모델 70	60.51	76.87	88.2	8.48	46.06	85.19	96.15
모델 71	49.28	67.59	83.68	14.28	33.12	83.41	94.95
모델 72	76.04	84.46	91.47	3.68	68.6	83.75	96.42
모델 73	45.86	64.74	70.57	8.38	29.2	86.13	96.61
모델 74	41.63	55.12	82.21	13.01	29.73	77.4	92.27
모델 75	61.49	78.4	88.68	5.7	46.56	84.85	96.33
모델 76	55.41	72.87	83.06	5.48	39.99	84.74	96.35
모델 77	26.45	45.69	55.23	11.49	9.47	50.93	60.81
모델 78	53.89	72.52	87.47	8.66	37.44	85.05	96.39
모델 79	55.57	72.39	83.32	10.18	40.74	84.24	96.15
모델 80	60.89	78.07	93.02	5.31	45.73	84.05	96.0
모델 81	60.87	79.12	92.61	8.56	44.76	85.63	96.35
모델 82	31.24	47.9	54.04	13.98	16.66	84.31	84.33
모델 83	54.06	71.23	81.59	5.34	38.91	84.48	96.16
모델 84	60.7	77.41	89.65	6.09	45.96	84.55	96.37
모델 85	47.72	66.28	84.16	19.05	31.34	84.52	95.45
모델 86	49.27	66.57	78.51	6.91	34.0	84.58	95.78
모델 87	41.07	61.9	68.68	17.93	22.69	87.3	96.39
모델 88	51.51	67.82	79.7	6.16	37.11	83.69	95.82
모델 89	56.09	73.4	85.11	5.14	40.82	84.79	96.37
모델 90	56.11	72.65	85.51	5.42	41.52	84.36	96.25
모델 91	57.79	75.37	88.3	5.65	42.26	84.85	96.11
모델 92	62.45	80.91	93.95	5.36	46.15	84.15	96.33
모델 93	58.91	76.08	89.52	6.93	43.75	84.52	96.18
모델 94	56.95	75.01	90.49	9.5	41.02	85.46	96.37
모델 95	60.83	78.9	92.49	5.54	44.89	84.55	96.19
lxmert-learning-cross-modality-encoder	60.33	77.16	89.59	5.69	45.47	84.53	96.35
모델 97	60.95	78.41	89.08	4.86	45.54	84.27	96.35
모델 98	56.38	74.84	91.71	6.32	40.09	83.76	95.43
모델 99	58.2	75.91	88.25	5.81	42.57	84.72	96.08
모델 100	56.28	73.73	86.86	5.78	40.87	84.2	96.01
모델 101	57.77	75.78	86.85	5.36	41.86	84.97	96.44
모델 102	58.72	76.4	89.58	6.58	43.11	84.68	96.21
모델 103	42.75	61.21	63.51	7.63	26.45	84.2	95.99
모델 104	73.33	79.68	77.02	2.46	67.73	83.7	96.36
모델 105	54.79	72.42	86.1	6.01	39.23	84.55	95.92
모델 106	52.3	68.46	84.36	12.54	38.04	85.2	96.2
모델 107	59.84	76.79	89.52	6.06	44.89	84.72	96.2
모델 108	60.18	76.84	89.77	5.65	45.48	84.6	96.37
모델 109	43.84	59.24	67.71	10.99	30.24	84.01	95.32
모델 110	59.72	77.97	89.43	6.25	43.61	84.89	96.55
모델 111	60.28	77.13	89.47	5.38	45.41	84.45	96.33
모델 112	58.12	76.39	88.01	5.65	42.0	84.8	96.06
모델 113	58.88	75.07	84.64	5.54	44.58	84.86	96.23
모델 114	60.01	76.77	89.17	6.28	45.21	84.46	96.35
모델 115	59.06	76.07	89.81	6.14	44.04	82.76	93.82
모델 116	58.42	77.39	90.29	7.86	41.67	84.53	95.57
모델 117	55.35	72.65	84.17	5.22	40.08	84.56	96.32
모델 118	53.85	68.44	80.2	5.84	40.97	85.19	96.28
모델 119	54.94	71.7	82.71	5.1	40.14	84.78	96.4
모델 120	73.81	80.8	91.76	1.7	67.64	83.9	96.73
모델 121	67.55	80.45	93.83	2.78	56.16	84.16	96.53
모델 122	57.07	73.77	84.68	4.7	42.33	84.81	96.48
모델 123	56.0	73.9	87.16	6.02	40.2	84.45	96.01
모델 124	57.01	74.78	87.74	6.06	41.32	84.25	96.03
모델 125	52.02	67.35	80.44	5.64	38.5	83.94	95.75
모델 126	74.03	82.12	89.0	1.29	66.89	83.58	96.76
모델 127	47.38	58.76	73.71	6.29	37.34	81.75	94.55