Add ScreenSpot-Pro evaluation result (GUI-Actor-2VL-2B)

#2
by merve HF Staff - opened
Files changed (1) hide show
  1. .eval_results/screenspot_pro.yaml +242 -0
.eval_results/screenspot_pro.yaml ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ - dataset:
2
+ id: likaixin/ScreenSpot-Pro
3
+ task_id: overall
4
+ value: 36.7
5
+ source:
6
+ url: https://gui-agent.github.io/grounding-leaderboard/
7
+ name: ScreenSpot-Pro Leaderboard
8
+ user: merve
9
+
10
+ - dataset:
11
+ id: likaixin/ScreenSpot-Pro
12
+ task_id: android_studio_macos
13
+ value: 33.8
14
+ source:
15
+ url: https://gui-agent.github.io/grounding-leaderboard/
16
+ name: ScreenSpot-Pro Leaderboard
17
+ user: merve
18
+
19
+ - dataset:
20
+ id: likaixin/ScreenSpot-Pro
21
+ task_id: autocad_windows
22
+ value: 8.8
23
+ source:
24
+ url: https://gui-agent.github.io/grounding-leaderboard/
25
+ name: ScreenSpot-Pro Leaderboard
26
+ user: merve
27
+
28
+ - dataset:
29
+ id: likaixin/ScreenSpot-Pro
30
+ task_id: blender_windows
31
+ value: 32.4
32
+ source:
33
+ url: https://gui-agent.github.io/grounding-leaderboard/
34
+ name: ScreenSpot-Pro Leaderboard
35
+ user: merve
36
+
37
+ - dataset:
38
+ id: likaixin/ScreenSpot-Pro
39
+ task_id: davinci_macos
40
+ value: 45.5
41
+ source:
42
+ url: https://gui-agent.github.io/grounding-leaderboard/
43
+ name: ScreenSpot-Pro Leaderboard
44
+ user: merve
45
+
46
+ - dataset:
47
+ id: likaixin/ScreenSpot-Pro
48
+ task_id: eviews_windows
49
+ value: 74.0
50
+ source:
51
+ url: https://gui-agent.github.io/grounding-leaderboard/
52
+ name: ScreenSpot-Pro Leaderboard
53
+ user: merve
54
+
55
+ - dataset:
56
+ id: likaixin/ScreenSpot-Pro
57
+ task_id: excel_macos
58
+ value: 37.5
59
+ source:
60
+ url: https://gui-agent.github.io/grounding-leaderboard/
61
+ name: ScreenSpot-Pro Leaderboard
62
+ user: merve
63
+
64
+ - dataset:
65
+ id: likaixin/ScreenSpot-Pro
66
+ task_id: fruitloops_windows
67
+ value: 33.3
68
+ source:
69
+ url: https://gui-agent.github.io/grounding-leaderboard/
70
+ name: ScreenSpot-Pro Leaderboard
71
+ user: merve
72
+
73
+ - dataset:
74
+ id: likaixin/ScreenSpot-Pro
75
+ task_id: illustrator_windows
76
+ value: 32.3
77
+ source:
78
+ url: https://gui-agent.github.io/grounding-leaderboard/
79
+ name: ScreenSpot-Pro Leaderboard
80
+ user: merve
81
+
82
+ - dataset:
83
+ id: likaixin/ScreenSpot-Pro
84
+ task_id: inventor_windows
85
+ value: 22.9
86
+ source:
87
+ url: https://gui-agent.github.io/grounding-leaderboard/
88
+ name: ScreenSpot-Pro Leaderboard
89
+ user: merve
90
+
91
+ - dataset:
92
+ id: likaixin/ScreenSpot-Pro
93
+ task_id: linux_common_linux
94
+ value: 28.0
95
+ source:
96
+ url: https://gui-agent.github.io/grounding-leaderboard/
97
+ name: ScreenSpot-Pro Leaderboard
98
+ user: merve
99
+
100
+ - dataset:
101
+ id: likaixin/ScreenSpot-Pro
102
+ task_id: macos_common_macos
103
+ value: 36.9
104
+ source:
105
+ url: https://gui-agent.github.io/grounding-leaderboard/
106
+ name: ScreenSpot-Pro Leaderboard
107
+ user: merve
108
+
109
+ - dataset:
110
+ id: likaixin/ScreenSpot-Pro
111
+ task_id: matlab_macos
112
+ value: 44.1
113
+ source:
114
+ url: https://gui-agent.github.io/grounding-leaderboard/
115
+ name: ScreenSpot-Pro Leaderboard
116
+ user: merve
117
+
118
+ - dataset:
119
+ id: likaixin/ScreenSpot-Pro
120
+ task_id: origin_windows
121
+ value: 12.9
122
+ source:
123
+ url: https://gui-agent.github.io/grounding-leaderboard/
124
+ name: ScreenSpot-Pro Leaderboard
125
+ user: merve
126
+
127
+ - dataset:
128
+ id: likaixin/ScreenSpot-Pro
129
+ task_id: photoshop_windows
130
+ value: 33.3
131
+ source:
132
+ url: https://gui-agent.github.io/grounding-leaderboard/
133
+ name: ScreenSpot-Pro Leaderboard
134
+ user: merve
135
+
136
+ - dataset:
137
+ id: likaixin/ScreenSpot-Pro
138
+ task_id: powerpoint_windows
139
+ value: 54.9
140
+ source:
141
+ url: https://gui-agent.github.io/grounding-leaderboard/
142
+ name: ScreenSpot-Pro Leaderboard
143
+ user: merve
144
+
145
+ - dataset:
146
+ id: likaixin/ScreenSpot-Pro
147
+ task_id: premiere_windows
148
+ value: 26.9
149
+ source:
150
+ url: https://gui-agent.github.io/grounding-leaderboard/
151
+ name: ScreenSpot-Pro Leaderboard
152
+ user: merve
153
+
154
+ - dataset:
155
+ id: likaixin/ScreenSpot-Pro
156
+ task_id: pycharm_macos
157
+ value: 30.8
158
+ source:
159
+ url: https://gui-agent.github.io/grounding-leaderboard/
160
+ name: ScreenSpot-Pro Leaderboard
161
+ user: merve
162
+
163
+ - dataset:
164
+ id: likaixin/ScreenSpot-Pro
165
+ task_id: quartus_windows
166
+ value: 20.0
167
+ source:
168
+ url: https://gui-agent.github.io/grounding-leaderboard/
169
+ name: ScreenSpot-Pro Leaderboard
170
+ user: merve
171
+
172
+ - dataset:
173
+ id: likaixin/ScreenSpot-Pro
174
+ task_id: solidworks_windows
175
+ value: 19.5
176
+ source:
177
+ url: https://gui-agent.github.io/grounding-leaderboard/
178
+ name: ScreenSpot-Pro Leaderboard
179
+ user: merve
180
+
181
+ - dataset:
182
+ id: likaixin/ScreenSpot-Pro
183
+ task_id: stata_windows
184
+ value: 22.4
185
+ source:
186
+ url: https://gui-agent.github.io/grounding-leaderboard/
187
+ name: ScreenSpot-Pro Leaderboard
188
+ user: merve
189
+
190
+ - dataset:
191
+ id: likaixin/ScreenSpot-Pro
192
+ task_id: unreal_engine_windows
193
+ value: 51.4
194
+ source:
195
+ url: https://gui-agent.github.io/grounding-leaderboard/
196
+ name: ScreenSpot-Pro Leaderboard
197
+ user: merve
198
+
199
+ - dataset:
200
+ id: likaixin/ScreenSpot-Pro
201
+ task_id: vivado_windows
202
+ value: 50.0
203
+ source:
204
+ url: https://gui-agent.github.io/grounding-leaderboard/
205
+ name: ScreenSpot-Pro Leaderboard
206
+ user: merve
207
+
208
+ - dataset:
209
+ id: likaixin/ScreenSpot-Pro
210
+ task_id: vmware_macos
211
+ value: 48.8
212
+ source:
213
+ url: https://gui-agent.github.io/grounding-leaderboard/
214
+ name: ScreenSpot-Pro Leaderboard
215
+ user: merve
216
+
217
+ - dataset:
218
+ id: likaixin/ScreenSpot-Pro
219
+ task_id: vscode_macos
220
+ value: 43.6
221
+ source:
222
+ url: https://gui-agent.github.io/grounding-leaderboard/
223
+ name: ScreenSpot-Pro Leaderboard
224
+ user: merve
225
+
226
+ - dataset:
227
+ id: likaixin/ScreenSpot-Pro
228
+ task_id: windows_common_windows
229
+ value: 27.2
230
+ source:
231
+ url: https://gui-agent.github.io/grounding-leaderboard/
232
+ name: ScreenSpot-Pro Leaderboard
233
+ user: merve
234
+
235
+ - dataset:
236
+ id: likaixin/ScreenSpot-Pro
237
+ task_id: word_macos
238
+ value: 65.5
239
+ source:
240
+ url: https://gui-agent.github.io/grounding-leaderboard/
241
+ name: ScreenSpot-Pro Leaderboard
242
+ user: merve