Hugging Face's logo

Spaces:
TIGER-Lab
/
MMEB-Leaderboard
Running

App Files Community
135

Upload RGE MMEB-V1 scores. (paper link: https://arxiv.org/abs/2511.16150)

#130
by lcxrocks - opened
base: refs/heads/main
←
from: refs/pr/130
Discussion Files changed
+972
-0
Files changed (4) hide show
  1. scores/RGE(Qwen2.5VL-3B).json +243 -0
  2. scores/RGE(Qwen2.5VL-7B).json +243 -0
  3. scores/RGE(Qwen2VL-2B).json +243 -0
  4. scores/RGE(Qwen2VL-7B).json +243 -0
scores/RGE(Qwen2.5VL-3B).json ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "model_name": "RGE(Qwen2.5VL-3B)",
4
+ "model_backbone": "Qwen2.5VL",
5
+ "model_size": "3B parameters",
6
+ "embedding_dimension": null,
7
+ "max_length_tokens": null,
8
+ "model_release_date": "2025-11-20",
9
+ "data_source": "Self-Reported",
10
+ "url": "https://huggingface.co/MCG-NJU/RGE",
11
+ "report_generated_date": "2026-05-11T15:49:30.024670"
12
+ },
13
+ "metrics": {
14
+ "image": {
15
+ "ImageNet-1K": {
16
+ "hit@1": 0.809,
17
+ "num_correct": 809,
18
+ "num_pred": 1000
19
+ },
20
+ "N24News": {
21
+ "hit@1": 0.805,
22
+ "num_correct": 805,
23
+ "num_pred": 1000
24
+ },
25
+ "HatefulMemes": {
26
+ "hit@1": 0.764,
27
+ "num_correct": 764,
28
+ "num_pred": 1000
29
+ },
30
+ "VOC2007": {
31
+ "hit@1": 0.878,
32
+ "num_correct": 878,
33
+ "num_pred": 1000
34
+ },
35
+ "SUN397": {
36
+ "hit@1": 0.769,
37
+ "num_correct": 769,
38
+ "num_pred": 1000
39
+ },
40
+ "Place365": {
41
+ "hit@1": 0.452,
42
+ "num_correct": 452,
43
+ "num_pred": 1000
44
+ },
45
+ "ImageNet-A": {
46
+ "hit@1": 0.359,
47
+ "num_correct": 359,
48
+ "num_pred": 1000
49
+ },
50
+ "ImageNet-R": {
51
+ "hit@1": 0.79,
52
+ "num_correct": 790,
53
+ "num_pred": 1000
54
+ },
55
+ "ObjectNet": {
56
+ "hit@1": 0.596,
57
+ "num_correct": 596,
58
+ "num_pred": 1000
59
+ },
60
+ "Country211": {
61
+ "hit@1": 0.219,
62
+ "num_correct": 219,
63
+ "num_pred": 1000
64
+ },
65
+ "OK-VQA": {
66
+ "hit@1": 0.692,
67
+ "num_correct": 692,
68
+ "num_pred": 1000
69
+ },
70
+ "A-OKVQA": {
71
+ "hit@1": 0.573,
72
+ "num_correct": 573,
73
+ "num_pred": 1000
74
+ },
75
+ "DocVQA": {
76
+ "hit@1": 0.934,
77
+ "num_correct": 934,
78
+ "num_pred": 1000
79
+ },
80
+ "InfographicsVQA": {
81
+ "hit@1": 0.739,
82
+ "num_correct": 739,
83
+ "num_pred": 1000
84
+ },
85
+ "ChartQA": {
86
+ "hit@1": 0.754,
87
+ "num_correct": 754,
88
+ "num_pred": 1000
89
+ },
90
+ "Visual7W": {
91
+ "hit@1": 0.554,
92
+ "num_correct": 554,
93
+ "num_pred": 1000
94
+ },
95
+ "ScienceQA": {
96
+ "hit@1": 0.524,
97
+ "num_correct": 524,
98
+ "num_pred": 1000
99
+ },
100
+ "VizWiz": {
101
+ "hit@1": 0.476,
102
+ "num_correct": 476,
103
+ "num_pred": 1000
104
+ },
105
+ "GQA": {
106
+ "hit@1": 0.741,
107
+ "num_correct": 741,
108
+ "num_pred": 1000
109
+ },
110
+ "TextVQA": {
111
+ "hit@1": 0.79,
112
+ "num_correct": 790,
113
+ "num_pred": 1000
114
+ },
115
+ "VisDial": {
116
+ "hit@1": 0.827,
117
+ "num_correct": 827,
118
+ "num_pred": 1000
119
+ },
120
+ "CIRR": {
121
+ "hit@1": 0.54,
122
+ "num_correct": 540,
123
+ "num_pred": 1000
124
+ },
125
+ "VisualNews_t2i": {
126
+ "hit@1": 0.756,
127
+ "num_correct": 756,
128
+ "num_pred": 1000
129
+ },
130
+ "VisualNews_i2t": {
131
+ "hit@1": 0.794,
132
+ "num_correct": 794,
133
+ "num_pred": 1000
134
+ },
135
+ "MSCOCO_t2i": {
136
+ "hit@1": 0.746,
137
+ "num_correct": 746,
138
+ "num_pred": 1000
139
+ },
140
+ "MSCOCO_i2t": {
141
+ "hit@1": 0.736,
142
+ "num_correct": 736,
143
+ "num_pred": 1000
144
+ },
145
+ "NIGHTS": {
146
+ "hit@1": 0.653,
147
+ "num_correct": 653,
148
+ "num_pred": 1000
149
+ },
150
+ "WebQA": {
151
+ "hit@1": 0.888,
152
+ "num_correct": 888,
153
+ "num_pred": 1000
154
+ },
155
+ "FashionIQ": {
156
+ "hit@1": 0.222,
157
+ "num_correct": 222,
158
+ "num_pred": 1000
159
+ },
160
+ "Wiki-SS-NQ": {
161
+ "hit@1": 0.657,
162
+ "num_correct": 657,
163
+ "num_pred": 1000
164
+ },
165
+ "OVEN": {
166
+ "hit@1": 0.69,
167
+ "num_correct": 690,
168
+ "num_pred": 1000
169
+ },
170
+ "EDIS": {
171
+ "hit@1": 0.913,
172
+ "num_correct": 913,
173
+ "num_pred": 1000
174
+ },
175
+ "MSCOCO": {
176
+ "hit@1": 0.841,
177
+ "num_correct": 841,
178
+ "num_pred": 1000
179
+ },
180
+ "RefCOCO": {
181
+ "hit@1": 0.946,
182
+ "num_correct": 946,
183
+ "num_pred": 1000
184
+ },
185
+ "RefCOCO-Matching": {
186
+ "hit@1": 0.928,
187
+ "num_correct": 928,
188
+ "num_pred": 1000
189
+ },
190
+ "Visual7W-Pointing": {
191
+ "hit@1": 0.888,
192
+ "num_correct": 888,
193
+ "num_pred": 1000
194
+ }
195
+ },
196
+ "video": {
197
+ "K700": "DIR_N/A",
198
+ "SmthSmthV2": "DIR_N/A",
199
+ "HMDB51": "DIR_N/A",
200
+ "UCF101": "DIR_N/A",
201
+ "Breakfast": "DIR_N/A",
202
+ "MVBench": "DIR_N/A",
203
+ "Video-MME": "DIR_N/A",
204
+ "NExTQA": "DIR_N/A",
205
+ "EgoSchema": "DIR_N/A",
206
+ "ActivityNetQA": "DIR_N/A",
207
+ "DiDeMo": "DIR_N/A",
208
+ "MSR-VTT": "DIR_N/A",
209
+ "MSVD": "DIR_N/A",
210
+ "VATEX": "DIR_N/A",
211
+ "YouCook2": "DIR_N/A",
212
+ "QVHighlight": "DIR_N/A",
213
+ "Charades-STA": "DIR_N/A",
214
+ "MomentSeeker": "DIR_N/A"
215
+ },
216
+ "visdoc": {
217
+ "ViDoRe_arxivqa": "DIR_N/A",
218
+ "ViDoRe_docvqa": "DIR_N/A",
219
+ "ViDoRe_infovqa": "DIR_N/A",
220
+ "ViDoRe_tabfquad": "DIR_N/A",
221
+ "ViDoRe_tatdqa": "DIR_N/A",
222
+ "ViDoRe_shiftproject": "DIR_N/A",
223
+ "ViDoRe_syntheticDocQA_artificial_intelligence": "DIR_N/A",
224
+ "ViDoRe_syntheticDocQA_energy": "DIR_N/A",
225
+ "ViDoRe_syntheticDocQA_government_reports": "DIR_N/A",
226
+ "ViDoRe_syntheticDocQA_healthcare_industry": "DIR_N/A",
227
+ "ViDoRe_esg_reports_human_labeled_v2": "DIR_N/A",
228
+ "ViDoRe_biomedical_lectures_v2_multilingual": "DIR_N/A",
229
+ "ViDoRe_economics_reports_v2_multilingual": "DIR_N/A",
230
+ "ViDoRe_esg_reports_v2_multilingual": "DIR_N/A",
231
+ "VisRAG_ArxivQA": "DIR_N/A",
232
+ "VisRAG_ChartQA": "DIR_N/A",
233
+ "VisRAG_MP-DocVQA": "DIR_N/A",
234
+ "VisRAG_SlideVQA": "DIR_N/A",
235
+ "VisRAG_InfoVQA": "DIR_N/A",
236
+ "VisRAG_PlotQA": "DIR_N/A",
237
+ "ViDoSeek-page-fixed": "DIR_N/A",
238
+ "ViDoSeek-doc": "DIR_N/A",
239
+ "MMLongBench-page-fixed": "DIR_N/A",
240
+ "MMLongBench-doc": "DIR_N/A"
241
+ }
242
+ }
243
+ }
scores/RGE(Qwen2.5VL-7B).json ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "model_name": "RGE(Qwen2.5VL-7B)",
4
+ "model_backbone": "Qwen2.5VL",
5
+ "model_size": "7B parameters",
6
+ "embedding_dimension": null,
7
+ "max_length_tokens": null,
8
+ "model_release_date": "2025-11-20",
9
+ "data_source": "Self-Reported",
10
+ "url": "https://huggingface.co/MCG-NJU/RGE",
11
+ "report_generated_date": "2026-05-11T15:49:31.062703"
12
+ },
13
+ "metrics": {
14
+ "image": {
15
+ "ImageNet-1K": {
16
+ "hit@1": 0.826,
17
+ "num_correct": 826,
18
+ "num_pred": 1000
19
+ },
20
+ "N24News": {
21
+ "hit@1": 0.804,
22
+ "num_correct": 804,
23
+ "num_pred": 1000
24
+ },
25
+ "HatefulMemes": {
26
+ "hit@1": 0.755,
27
+ "num_correct": 755,
28
+ "num_pred": 1000
29
+ },
30
+ "VOC2007": {
31
+ "hit@1": 0.873,
32
+ "num_correct": 873,
33
+ "num_pred": 1000
34
+ },
35
+ "SUN397": {
36
+ "hit@1": 0.806,
37
+ "num_correct": 806,
38
+ "num_pred": 1000
39
+ },
40
+ "Place365": {
41
+ "hit@1": 0.471,
42
+ "num_correct": 471,
43
+ "num_pred": 1000
44
+ },
45
+ "ImageNet-A": {
46
+ "hit@1": 0.486,
47
+ "num_correct": 486,
48
+ "num_pred": 1000
49
+ },
50
+ "ImageNet-R": {
51
+ "hit@1": 0.805,
52
+ "num_correct": 805,
53
+ "num_pred": 1000
54
+ },
55
+ "ObjectNet": {
56
+ "hit@1": 0.659,
57
+ "num_correct": 659,
58
+ "num_pred": 1000
59
+ },
60
+ "Country211": {
61
+ "hit@1": 0.243,
62
+ "num_correct": 243,
63
+ "num_pred": 1000
64
+ },
65
+ "OK-VQA": {
66
+ "hit@1": 0.731,
67
+ "num_correct": 731,
68
+ "num_pred": 1000
69
+ },
70
+ "A-OKVQA": {
71
+ "hit@1": 0.615,
72
+ "num_correct": 615,
73
+ "num_pred": 1000
74
+ },
75
+ "DocVQA": {
76
+ "hit@1": 0.944,
77
+ "num_correct": 944,
78
+ "num_pred": 1000
79
+ },
80
+ "InfographicsVQA": {
81
+ "hit@1": 0.789,
82
+ "num_correct": 789,
83
+ "num_pred": 1000
84
+ },
85
+ "ChartQA": {
86
+ "hit@1": 0.805,
87
+ "num_correct": 805,
88
+ "num_pred": 1000
89
+ },
90
+ "Visual7W": {
91
+ "hit@1": 0.561,
92
+ "num_correct": 561,
93
+ "num_pred": 1000
94
+ },
95
+ "ScienceQA": {
96
+ "hit@1": 0.577,
97
+ "num_correct": 577,
98
+ "num_pred": 1000
99
+ },
100
+ "VizWiz": {
101
+ "hit@1": 0.476,
102
+ "num_correct": 476,
103
+ "num_pred": 1000
104
+ },
105
+ "GQA": {
106
+ "hit@1": 0.771,
107
+ "num_correct": 771,
108
+ "num_pred": 1000
109
+ },
110
+ "TextVQA": {
111
+ "hit@1": 0.83,
112
+ "num_correct": 830,
113
+ "num_pred": 1000
114
+ },
115
+ "VisDial": {
116
+ "hit@1": 0.837,
117
+ "num_correct": 837,
118
+ "num_pred": 1000
119
+ },
120
+ "CIRR": {
121
+ "hit@1": 0.575,
122
+ "num_correct": 575,
123
+ "num_pred": 1000
124
+ },
125
+ "VisualNews_t2i": {
126
+ "hit@1": 0.764,
127
+ "num_correct": 764,
128
+ "num_pred": 1000
129
+ },
130
+ "VisualNews_i2t": {
131
+ "hit@1": 0.818,
132
+ "num_correct": 818,
133
+ "num_pred": 1000
134
+ },
135
+ "MSCOCO_t2i": {
136
+ "hit@1": 0.777,
137
+ "num_correct": 777,
138
+ "num_pred": 1000
139
+ },
140
+ "MSCOCO_i2t": {
141
+ "hit@1": 0.746,
142
+ "num_correct": 746,
143
+ "num_pred": 1000
144
+ },
145
+ "NIGHTS": {
146
+ "hit@1": 0.68,
147
+ "num_correct": 680,
148
+ "num_pred": 1000
149
+ },
150
+ "WebQA": {
151
+ "hit@1": 0.9,
152
+ "num_correct": 900,
153
+ "num_pred": 1000
154
+ },
155
+ "FashionIQ": {
156
+ "hit@1": 0.281,
157
+ "num_correct": 281,
158
+ "num_pred": 1000
159
+ },
160
+ "Wiki-SS-NQ": {
161
+ "hit@1": 0.681,
162
+ "num_correct": 681,
163
+ "num_pred": 1000
164
+ },
165
+ "OVEN": {
166
+ "hit@1": 0.748,
167
+ "num_correct": 748,
168
+ "num_pred": 1000
169
+ },
170
+ "EDIS": {
171
+ "hit@1": 0.939,
172
+ "num_correct": 939,
173
+ "num_pred": 1000
174
+ },
175
+ "MSCOCO": {
176
+ "hit@1": 0.829,
177
+ "num_correct": 829,
178
+ "num_pred": 1000
179
+ },
180
+ "RefCOCO": {
181
+ "hit@1": 0.935,
182
+ "num_correct": 935,
183
+ "num_pred": 1000
184
+ },
185
+ "RefCOCO-Matching": {
186
+ "hit@1": 0.929,
187
+ "num_correct": 929,
188
+ "num_pred": 1000
189
+ },
190
+ "Visual7W-Pointing": {
191
+ "hit@1": 0.881,
192
+ "num_correct": 881,
193
+ "num_pred": 1000
194
+ }
195
+ },
196
+ "video": {
197
+ "K700": "DIR_N/A",
198
+ "SmthSmthV2": "DIR_N/A",
199
+ "HMDB51": "DIR_N/A",
200
+ "UCF101": "DIR_N/A",
201
+ "Breakfast": "DIR_N/A",
202
+ "MVBench": "DIR_N/A",
203
+ "Video-MME": "DIR_N/A",
204
+ "NExTQA": "DIR_N/A",
205
+ "EgoSchema": "DIR_N/A",
206
+ "ActivityNetQA": "DIR_N/A",
207
+ "DiDeMo": "DIR_N/A",
208
+ "MSR-VTT": "DIR_N/A",
209
+ "MSVD": "DIR_N/A",
210
+ "VATEX": "DIR_N/A",
211
+ "YouCook2": "DIR_N/A",
212
+ "QVHighlight": "DIR_N/A",
213
+ "Charades-STA": "DIR_N/A",
214
+ "MomentSeeker": "DIR_N/A"
215
+ },
216
+ "visdoc": {
217
+ "ViDoRe_arxivqa": "DIR_N/A",
218
+ "ViDoRe_docvqa": "DIR_N/A",
219
+ "ViDoRe_infovqa": "DIR_N/A",
220
+ "ViDoRe_tabfquad": "DIR_N/A",
221
+ "ViDoRe_tatdqa": "DIR_N/A",
222
+ "ViDoRe_shiftproject": "DIR_N/A",
223
+ "ViDoRe_syntheticDocQA_artificial_intelligence": "DIR_N/A",
224
+ "ViDoRe_syntheticDocQA_energy": "DIR_N/A",
225
+ "ViDoRe_syntheticDocQA_government_reports": "DIR_N/A",
226
+ "ViDoRe_syntheticDocQA_healthcare_industry": "DIR_N/A",
227
+ "ViDoRe_esg_reports_human_labeled_v2": "DIR_N/A",
228
+ "ViDoRe_biomedical_lectures_v2_multilingual": "DIR_N/A",
229
+ "ViDoRe_economics_reports_v2_multilingual": "DIR_N/A",
230
+ "ViDoRe_esg_reports_v2_multilingual": "DIR_N/A",
231
+ "VisRAG_ArxivQA": "DIR_N/A",
232
+ "VisRAG_ChartQA": "DIR_N/A",
233
+ "VisRAG_MP-DocVQA": "DIR_N/A",
234
+ "VisRAG_SlideVQA": "DIR_N/A",
235
+ "VisRAG_InfoVQA": "DIR_N/A",
236
+ "VisRAG_PlotQA": "DIR_N/A",
237
+ "ViDoSeek-page-fixed": "DIR_N/A",
238
+ "ViDoSeek-doc": "DIR_N/A",
239
+ "MMLongBench-page-fixed": "DIR_N/A",
240
+ "MMLongBench-doc": "DIR_N/A"
241
+ }
242
+ }
243
+ }
scores/RGE(Qwen2VL-2B).json ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "model_name": "RGE(Qwen2VL-2B)",
4
+ "model_backbone": "Qwen2VL",
5
+ "model_size": "2B parameters",
6
+ "embedding_dimension": null,
7
+ "max_length_tokens": null,
8
+ "model_release_date": "2025-11-20",
9
+ "data_source": "Self-Reported",
10
+ "url": "https://huggingface.co/MCG-NJU/RGE",
11
+ "report_generated_date": "2026-05-11T15:49:31.956978"
12
+ },
13
+ "metrics": {
14
+ "image": {
15
+ "ImageNet-1K": {
16
+ "hit@1": 0.834,
17
+ "num_correct": 834,
18
+ "num_pred": 1000
19
+ },
20
+ "N24News": {
21
+ "hit@1": 0.797,
22
+ "num_correct": 797,
23
+ "num_pred": 1000
24
+ },
25
+ "HatefulMemes": {
26
+ "hit@1": 0.68,
27
+ "num_correct": 680,
28
+ "num_pred": 1000
29
+ },
30
+ "VOC2007": {
31
+ "hit@1": 0.897,
32
+ "num_correct": 897,
33
+ "num_pred": 1000
34
+ },
35
+ "SUN397": {
36
+ "hit@1": 0.799,
37
+ "num_correct": 799,
38
+ "num_pred": 1000
39
+ },
40
+ "Place365": {
41
+ "hit@1": 0.461,
42
+ "num_correct": 461,
43
+ "num_pred": 1000
44
+ },
45
+ "ImageNet-A": {
46
+ "hit@1": 0.458,
47
+ "num_correct": 458,
48
+ "num_pred": 1000
49
+ },
50
+ "ImageNet-R": {
51
+ "hit@1": 0.887,
52
+ "num_correct": 887,
53
+ "num_pred": 1000
54
+ },
55
+ "ObjectNet": {
56
+ "hit@1": 0.696,
57
+ "num_correct": 696,
58
+ "num_pred": 1000
59
+ },
60
+ "Country211": {
61
+ "hit@1": 0.21,
62
+ "num_correct": 210,
63
+ "num_pred": 1000
64
+ },
65
+ "OK-VQA": {
66
+ "hit@1": 0.647,
67
+ "num_correct": 647,
68
+ "num_pred": 1000
69
+ },
70
+ "A-OKVQA": {
71
+ "hit@1": 0.54,
72
+ "num_correct": 540,
73
+ "num_pred": 1000
74
+ },
75
+ "DocVQA": {
76
+ "hit@1": 0.93,
77
+ "num_correct": 930,
78
+ "num_pred": 1000
79
+ },
80
+ "InfographicsVQA": {
81
+ "hit@1": 0.671,
82
+ "num_correct": 671,
83
+ "num_pred": 1000
84
+ },
85
+ "ChartQA": {
86
+ "hit@1": 0.678,
87
+ "num_correct": 678,
88
+ "num_pred": 1000
89
+ },
90
+ "Visual7W": {
91
+ "hit@1": 0.54,
92
+ "num_correct": 540,
93
+ "num_pred": 1000
94
+ },
95
+ "ScienceQA": {
96
+ "hit@1": 0.438,
97
+ "num_correct": 438,
98
+ "num_pred": 1000
99
+ },
100
+ "VizWiz": {
101
+ "hit@1": 0.464,
102
+ "num_correct": 464,
103
+ "num_pred": 1000
104
+ },
105
+ "GQA": {
106
+ "hit@1": 0.673,
107
+ "num_correct": 673,
108
+ "num_pred": 1000
109
+ },
110
+ "TextVQA": {
111
+ "hit@1": 0.75,
112
+ "num_correct": 750,
113
+ "num_pred": 1000
114
+ },
115
+ "VisDial": {
116
+ "hit@1": 0.798,
117
+ "num_correct": 798,
118
+ "num_pred": 1000
119
+ },
120
+ "CIRR": {
121
+ "hit@1": 0.554,
122
+ "num_correct": 554,
123
+ "num_pred": 1000
124
+ },
125
+ "VisualNews_t2i": {
126
+ "hit@1": 0.775,
127
+ "num_correct": 775,
128
+ "num_pred": 1000
129
+ },
130
+ "VisualNews_i2t": {
131
+ "hit@1": 0.81,
132
+ "num_correct": 810,
133
+ "num_pred": 1000
134
+ },
135
+ "MSCOCO_t2i": {
136
+ "hit@1": 0.767,
137
+ "num_correct": 767,
138
+ "num_pred": 1000
139
+ },
140
+ "MSCOCO_i2t": {
141
+ "hit@1": 0.732,
142
+ "num_correct": 732,
143
+ "num_pred": 1000
144
+ },
145
+ "NIGHTS": {
146
+ "hit@1": 0.665,
147
+ "num_correct": 665,
148
+ "num_pred": 1000
149
+ },
150
+ "WebQA": {
151
+ "hit@1": 0.884,
152
+ "num_correct": 884,
153
+ "num_pred": 1000
154
+ },
155
+ "FashionIQ": {
156
+ "hit@1": 0.238,
157
+ "num_correct": 238,
158
+ "num_pred": 1000
159
+ },
160
+ "Wiki-SS-NQ": {
161
+ "hit@1": 0.629,
162
+ "num_correct": 629,
163
+ "num_pred": 1000
164
+ },
165
+ "OVEN": {
166
+ "hit@1": 0.696,
167
+ "num_correct": 696,
168
+ "num_pred": 1000
169
+ },
170
+ "EDIS": {
171
+ "hit@1": 0.901,
172
+ "num_correct": 901,
173
+ "num_pred": 1000
174
+ },
175
+ "MSCOCO": {
176
+ "hit@1": 0.868,
177
+ "num_correct": 868,
178
+ "num_pred": 1000
179
+ },
180
+ "RefCOCO": {
181
+ "hit@1": 0.881,
182
+ "num_correct": 881,
183
+ "num_pred": 1000
184
+ },
185
+ "RefCOCO-Matching": {
186
+ "hit@1": 0.868,
187
+ "num_correct": 868,
188
+ "num_pred": 1000
189
+ },
190
+ "Visual7W-Pointing": {
191
+ "hit@1": 0.893,
192
+ "num_correct": 893,
193
+ "num_pred": 1000
194
+ }
195
+ },
196
+ "video": {
197
+ "K700": "DIR_N/A",
198
+ "SmthSmthV2": "DIR_N/A",
199
+ "HMDB51": "DIR_N/A",
200
+ "UCF101": "DIR_N/A",
201
+ "Breakfast": "DIR_N/A",
202
+ "MVBench": "DIR_N/A",
203
+ "Video-MME": "DIR_N/A",
204
+ "NExTQA": "DIR_N/A",
205
+ "EgoSchema": "DIR_N/A",
206
+ "ActivityNetQA": "DIR_N/A",
207
+ "DiDeMo": "DIR_N/A",
208
+ "MSR-VTT": "DIR_N/A",
209
+ "MSVD": "DIR_N/A",
210
+ "VATEX": "DIR_N/A",
211
+ "YouCook2": "DIR_N/A",
212
+ "QVHighlight": "DIR_N/A",
213
+ "Charades-STA": "DIR_N/A",
214
+ "MomentSeeker": "DIR_N/A"
215
+ },
216
+ "visdoc": {
217
+ "ViDoRe_arxivqa": "DIR_N/A",
218
+ "ViDoRe_docvqa": "DIR_N/A",
219
+ "ViDoRe_infovqa": "DIR_N/A",
220
+ "ViDoRe_tabfquad": "DIR_N/A",
221
+ "ViDoRe_tatdqa": "DIR_N/A",
222
+ "ViDoRe_shiftproject": "DIR_N/A",
223
+ "ViDoRe_syntheticDocQA_artificial_intelligence": "DIR_N/A",
224
+ "ViDoRe_syntheticDocQA_energy": "DIR_N/A",
225
+ "ViDoRe_syntheticDocQA_government_reports": "DIR_N/A",
226
+ "ViDoRe_syntheticDocQA_healthcare_industry": "DIR_N/A",
227
+ "ViDoRe_esg_reports_human_labeled_v2": "DIR_N/A",
228
+ "ViDoRe_biomedical_lectures_v2_multilingual": "DIR_N/A",
229
+ "ViDoRe_economics_reports_v2_multilingual": "DIR_N/A",
230
+ "ViDoRe_esg_reports_v2_multilingual": "DIR_N/A",
231
+ "VisRAG_ArxivQA": "DIR_N/A",
232
+ "VisRAG_ChartQA": "DIR_N/A",
233
+ "VisRAG_MP-DocVQA": "DIR_N/A",
234
+ "VisRAG_SlideVQA": "DIR_N/A",
235
+ "VisRAG_InfoVQA": "DIR_N/A",
236
+ "VisRAG_PlotQA": "DIR_N/A",
237
+ "ViDoSeek-page-fixed": "DIR_N/A",
238
+ "ViDoSeek-doc": "DIR_N/A",
239
+ "MMLongBench-page-fixed": "DIR_N/A",
240
+ "MMLongBench-doc": "DIR_N/A"
241
+ }
242
+ }
243
+ }
scores/RGE(Qwen2VL-7B).json ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "model_name": "RGE(Qwen2VL-7B)",
4
+ "model_backbone": "Qwen2VL",
5
+ "model_size": "7B parameters",
6
+ "embedding_dimension": null,
7
+ "max_length_tokens": null,
8
+ "model_release_date": "2025-11-20",
9
+ "data_source": "Self-Reported",
10
+ "url": "https://huggingface.co/MCG-NJU/RGE",
11
+ "report_generated_date": "2026-05-11T15:49:32.810768"
12
+ },
13
+ "metrics": {
14
+ "image": {
15
+ "ImageNet-1K": {
16
+ "hit@1": 0.827,
17
+ "num_correct": 827,
18
+ "num_pred": 1000
19
+ },
20
+ "N24News": {
21
+ "hit@1": 0.818,
22
+ "num_correct": 818,
23
+ "num_pred": 1000
24
+ },
25
+ "HatefulMemes": {
26
+ "hit@1": 0.769,
27
+ "num_correct": 769,
28
+ "num_pred": 1000
29
+ },
30
+ "VOC2007": {
31
+ "hit@1": 0.917,
32
+ "num_correct": 917,
33
+ "num_pred": 1000
34
+ },
35
+ "SUN397": {
36
+ "hit@1": 0.818,
37
+ "num_correct": 818,
38
+ "num_pred": 1000
39
+ },
40
+ "Place365": {
41
+ "hit@1": 0.485,
42
+ "num_correct": 485,
43
+ "num_pred": 1000
44
+ },
45
+ "ImageNet-A": {
46
+ "hit@1": 0.555,
47
+ "num_correct": 555,
48
+ "num_pred": 1000
49
+ },
50
+ "ImageNet-R": {
51
+ "hit@1": 0.887,
52
+ "num_correct": 887,
53
+ "num_pred": 1000
54
+ },
55
+ "ObjectNet": {
56
+ "hit@1": 0.724,
57
+ "num_correct": 724,
58
+ "num_pred": 1000
59
+ },
60
+ "Country211": {
61
+ "hit@1": 0.299,
62
+ "num_correct": 299,
63
+ "num_pred": 1000
64
+ },
65
+ "OK-VQA": {
66
+ "hit@1": 0.746,
67
+ "num_correct": 746,
68
+ "num_pred": 1000
69
+ },
70
+ "A-OKVQA": {
71
+ "hit@1": 0.609,
72
+ "num_correct": 609,
73
+ "num_pred": 1000
74
+ },
75
+ "DocVQA": {
76
+ "hit@1": 0.946,
77
+ "num_correct": 946,
78
+ "num_pred": 1000
79
+ },
80
+ "InfographicsVQA": {
81
+ "hit@1": 0.767,
82
+ "num_correct": 767,
83
+ "num_pred": 1000
84
+ },
85
+ "ChartQA": {
86
+ "hit@1": 0.78,
87
+ "num_correct": 780,
88
+ "num_pred": 1000
89
+ },
90
+ "Visual7W": {
91
+ "hit@1": 0.558,
92
+ "num_correct": 558,
93
+ "num_pred": 1000
94
+ },
95
+ "ScienceQA": {
96
+ "hit@1": 0.551,
97
+ "num_correct": 551,
98
+ "num_pred": 1000
99
+ },
100
+ "VizWiz": {
101
+ "hit@1": 0.533,
102
+ "num_correct": 533,
103
+ "num_pred": 1000
104
+ },
105
+ "GQA": {
106
+ "hit@1": 0.736,
107
+ "num_correct": 736,
108
+ "num_pred": 1000
109
+ },
110
+ "TextVQA": {
111
+ "hit@1": 0.838,
112
+ "num_correct": 838,
113
+ "num_pred": 1000
114
+ },
115
+ "VisDial": {
116
+ "hit@1": 0.827,
117
+ "num_correct": 827,
118
+ "num_pred": 1000
119
+ },
120
+ "CIRR": {
121
+ "hit@1": 0.601,
122
+ "num_correct": 601,
123
+ "num_pred": 1000
124
+ },
125
+ "VisualNews_t2i": {
126
+ "hit@1": 0.818,
127
+ "num_correct": 818,
128
+ "num_pred": 1000
129
+ },
130
+ "VisualNews_i2t": {
131
+ "hit@1": 0.864,
132
+ "num_correct": 864,
133
+ "num_pred": 1000
134
+ },
135
+ "MSCOCO_t2i": {
136
+ "hit@1": 0.794,
137
+ "num_correct": 794,
138
+ "num_pred": 1000
139
+ },
140
+ "MSCOCO_i2t": {
141
+ "hit@1": 0.774,
142
+ "num_correct": 774,
143
+ "num_pred": 1000
144
+ },
145
+ "NIGHTS": {
146
+ "hit@1": 0.683,
147
+ "num_correct": 683,
148
+ "num_pred": 1000
149
+ },
150
+ "WebQA": {
151
+ "hit@1": 0.894,
152
+ "num_correct": 894,
153
+ "num_pred": 1000
154
+ },
155
+ "FashionIQ": {
156
+ "hit@1": 0.276,
157
+ "num_correct": 276,
158
+ "num_pred": 1000
159
+ },
160
+ "Wiki-SS-NQ": {
161
+ "hit@1": 0.678,
162
+ "num_correct": 678,
163
+ "num_pred": 1000
164
+ },
165
+ "OVEN": {
166
+ "hit@1": 0.746,
167
+ "num_correct": 746,
168
+ "num_pred": 1000
169
+ },
170
+ "EDIS": {
171
+ "hit@1": 0.956,
172
+ "num_correct": 956,
173
+ "num_pred": 1000
174
+ },
175
+ "MSCOCO": {
176
+ "hit@1": 0.869,
177
+ "num_correct": 869,
178
+ "num_pred": 1000
179
+ },
180
+ "RefCOCO": {
181
+ "hit@1": 0.957,
182
+ "num_correct": 957,
183
+ "num_pred": 1000
184
+ },
185
+ "RefCOCO-Matching": {
186
+ "hit@1": 0.929,
187
+ "num_correct": 929,
188
+ "num_pred": 1000
189
+ },
190
+ "Visual7W-Pointing": {
191
+ "hit@1": 0.925,
192
+ "num_correct": 925,
193
+ "num_pred": 1000
194
+ }
195
+ },
196
+ "video": {
197
+ "K700": "DIR_N/A",
198
+ "SmthSmthV2": "DIR_N/A",
199
+ "HMDB51": "DIR_N/A",
200
+ "UCF101": "DIR_N/A",
201
+ "Breakfast": "DIR_N/A",
202
+ "MVBench": "DIR_N/A",
203
+ "Video-MME": "DIR_N/A",
204
+ "NExTQA": "DIR_N/A",
205
+ "EgoSchema": "DIR_N/A",
206
+ "ActivityNetQA": "DIR_N/A",
207
+ "DiDeMo": "DIR_N/A",
208
+ "MSR-VTT": "DIR_N/A",
209
+ "MSVD": "DIR_N/A",
210
+ "VATEX": "DIR_N/A",
211
+ "YouCook2": "DIR_N/A",
212
+ "QVHighlight": "DIR_N/A",
213
+ "Charades-STA": "DIR_N/A",
214
+ "MomentSeeker": "DIR_N/A"
215
+ },
216
+ "visdoc": {
217
+ "ViDoRe_arxivqa": "DIR_N/A",
218
+ "ViDoRe_docvqa": "DIR_N/A",
219
+ "ViDoRe_infovqa": "DIR_N/A",
220
+ "ViDoRe_tabfquad": "DIR_N/A",
221
+ "ViDoRe_tatdqa": "DIR_N/A",
222
+ "ViDoRe_shiftproject": "DIR_N/A",
223
+ "ViDoRe_syntheticDocQA_artificial_intelligence": "DIR_N/A",
224
+ "ViDoRe_syntheticDocQA_energy": "DIR_N/A",
225
+ "ViDoRe_syntheticDocQA_government_reports": "DIR_N/A",
226
+ "ViDoRe_syntheticDocQA_healthcare_industry": "DIR_N/A",
227
+ "ViDoRe_esg_reports_human_labeled_v2": "DIR_N/A",
228
+ "ViDoRe_biomedical_lectures_v2_multilingual": "DIR_N/A",
229
+ "ViDoRe_economics_reports_v2_multilingual": "DIR_N/A",
230
+ "ViDoRe_esg_reports_v2_multilingual": "DIR_N/A",
231
+ "VisRAG_ArxivQA": "DIR_N/A",
232
+ "VisRAG_ChartQA": "DIR_N/A",
233
+ "VisRAG_MP-DocVQA": "DIR_N/A",
234
+ "VisRAG_SlideVQA": "DIR_N/A",
235
+ "VisRAG_InfoVQA": "DIR_N/A",
236
+ "VisRAG_PlotQA": "DIR_N/A",
237
+ "ViDoSeek-page-fixed": "DIR_N/A",
238
+ "ViDoSeek-doc": "DIR_N/A",
239
+ "MMLongBench-page-fixed": "DIR_N/A",
240
+ "MMLongBench-doc": "DIR_N/A"
241
+ }
242
+ }
243
+ }