Dataset Viewer
image_path
stringlengths 28
84
| prompt
stringlengths 11
3.06k
| normalized_bbox
listlengths 4
4
| images
images listlengths 1
1
| easyr1_prompt
stringlengths 268
3.32k
| bbox
listlengths 4
4
| messages
listlengths 3
3
| original_image_width
int64 196
15k
| image_width
int64 196
3.75k
| image_height
int64 112
5.38k
|
|---|---|---|---|---|---|---|---|---|---|
doc_images/AmHHgw-Nep9dv3yPkTdG-MxD1Wka_1920x1080_CHAR_53_176.png
|
<image>
Given the following text: "The quality of the data and annotation upper-bounds the quality of a downstream model. While there exist large text corpora and image-text pairs, high-quality video-text data is much harder to collect. First of all, manual labeling is more time-consuming, as it requires an annotator to watch an entire video. Second, videos have a temporal dimension, consisting of several scenes stacked together, and showing multiple actions. Accordingly, to establish a video dataset with high-quality captions, we propose an automatic approach leveraging multimodal inputs, such as textual video description, subtitles, and individual video frames. Specifically, we curate 3.8M high-resolution videos from the publicly available HD-VILA-100M dataset. We then split them into semantically consistent video clips, and apply multiple cross-modality teacher models to obtain captions for each video. Next, we finetune a retrieval model on a small subset where the best caption of each video is manually selected and then employ the model in the whole dataset to select the best caption as the annotation. In " , find the text in the document and click the 5th character "s" in the text.
|
[
0.43854166666666666,
0.6648148148148149,
0.44375,
0.674074074074074
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "The quality of the data and annotation upper-bounds the quality of a downstream model. While there exist large text corpora and image-text pairs, high-quality video-text data is much harder to collect. First of all, manual labeling is more time-consuming, as it requires an annotator to watch an entire video. Second, videos have a temporal dimension, consisting of several scenes stacked together, and showing multiple actions. Accordingly, to establish a video dataset with high-quality captions, we propose an automatic approach leveraging multimodal inputs, such as textual video description, subtitles, and individual video frames. Specifically, we curate 3.8M high-resolution videos from the publicly available HD-VILA-100M dataset. We then split them into semantically consistent video clips, and apply multiple cross-modality teacher models to obtain captions for each video. Next, we finetune a retrieval model on a small subset where the best caption of each video is manually selected and then employ the model in the whole dataset to select the best caption as the annotation. In " , find the text in the document and click the 5th character "s" in the text.
|
[
847,
725,
857,
736
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"The quality of the data and annotation upper-bounds the quality of a downstream model. While there exist large text corpora and image-text pairs, high-quality video-text data is much harder to collect. First of all, manual labeling is more time-consuming, as it requires an annotator to watch an entire video. Second, videos have a temporal dimension, consisting of several scenes stacked together, and showing multiple actions. Accordingly, to establish a video dataset with high-quality captions, we propose an automatic approach leveraging multimodal inputs, such as textual video description, subtitles, and individual video frames. Specifically, we curate 3.8M high-resolution videos from the publicly available HD-VILA-100M dataset. We then split them into semantically consistent video clips, and apply multiple cross-modality teacher models to obtain captions for each video. Next, we finetune a retrieval model on a small subset where the best caption of each video is manually selected and then employ the model in the whole dataset to select the best caption as the annotation. In \" , find the text in the document and click the 5th character \"s\" in the text.",
"role": "user"
},
{
"content": "(852,730)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
uground-images/uground_555198.jpg
|
<image>
Point to Purple background, simple black musical note with looped head, under "Privacy Policy | Refund Policy | Shipping Policy"
|
[
0.5046246647834778,
0.713546872138977,
0.5342265963554382,
0.749249279499054
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Point to Purple background, simple black musical note with looped head, under "Privacy Policy | Refund Policy | Shipping Policy"
|
[
876,
1018,
927,
1069
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nPoint to Purple background, simple black musical note with looped head, under \"Privacy Policy | Refund Policy | Shipping Policy\"",
"role": "user"
},
{
"content": "(901,1043)",
"role": "assistant"
}
] | 1,731
| 1,736
| 1,428
|
|
showui-web-images/showui_web_001850.jpg
|
<image>
Select Hotels in South Lake Tahoe, United States of America
|
[
0.42895641922950745,
0.4844900965690613,
0.5570362210273743,
0.5069918036460876
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select Hotels in South Lake Tahoe, United States of America
|
[
528,
325,
686,
340
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect Hotels in South Lake Tahoe, United States of America",
"role": "user"
},
{
"content": "(607,332)",
"role": "assistant"
}
] | 1,219
| 1,232
| 672
|
|
showui-desktop-images/showui_desktop_004480.jpg
|
<image>
Navigate to cube_root
|
[
0.3894040286540985,
0.48513394594192505,
0.42484429478645325,
0.5313547849655151
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Navigate to cube_root
|
[
981,
760,
1070,
833
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nNavigate to cube_root",
"role": "user"
},
{
"content": "(1025,796)",
"role": "assistant"
}
] | 2,533
| 2,520
| 1,568
|
|
doc_images/AmHHgw-Nep9dv1qwyRqBwq-G7dOS_3840x2160_CHAR_6_104.png
|
<image>
Given the following text: " high frequency words, some more prevalent than others. Dolch’s included the following sight words on his primer word list." , find the text in the document and click the 7th character "s" in the text.
|
[
0.28932291666666665,
0.1037037037037037,
0.29192708333333334,
0.10833333333333334
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: " high frequency words, some more prevalent than others. Dolch’s included the following sight words on his primer word list." , find the text in the document and click the 7th character "s" in the text.
|
[
769,
153,
776,
160
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \" high frequency words, some more prevalent than others. Dolch’s included the following sight words on his primer word list.\" , find the text in the document and click the 7th character \"s\" in the text.",
"role": "user"
},
{
"content": "(772,156)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
autogui-images/autogui_2210935_69237a49.png
|
<image>
In this webpage capture, I will describe a certain element. Please locate it for me (with point). This element allows users to reorganize the list of messages according to a thread-based sorting criterion.
|
[
0.1887018382549286,
0.5025158524513245,
0.22677847743034363,
0.5321590304374695
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
In this webpage capture, I will describe a certain element. Please locate it for me (with point). This element allows users to reorganize the list of messages according to a thread-based sorting criterion.
|
[
142,
211,
171,
223
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nIn this webpage capture, I will describe a certain element. Please locate it for me (with point). This element allows users to reorganize the list of messages according to a thread-based sorting criterion.",
"role": "user"
},
{
"content": "(156,217)",
"role": "assistant"
}
] | 756
| 756
| 420
|
|
autogui-images/autogui_2501592_8da1dde6.png
|
<image>
Using the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element resets the angle of the map view to its default position, helping users return the map to its initial orientation after adjusting the pitch.
|
[
0.9474844932556152,
0.4353940784931183,
0.9808886647224426,
0.48813381791114807
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Using the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element resets the angle of the map view to its default position, helping users return the map to its initial orientation after adjusting the pitch.
|
[
530,
134,
549,
150
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nUsing the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element resets the angle of the map view to its default position, helping users return the map to its initial orientation after adjusting the pitch.",
"role": "user"
},
{
"content": "(539,142)",
"role": "assistant"
}
] | 560
| 560
| 308
|
|
cell_right_bottom_corner/images/1920x1080_cell_R58.png
|
<image>
Series fill point of cell R58
|
[
0.7727743271221532,
0.9423076923076923,
0.7779503105590062,
0.9514652014652014
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Series fill point of cell R58
|
[
1493,
1029,
1503,
1039
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSeries fill point of cell R58",
"role": "user"
},
{
"content": "(1498,1034)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
osatlas-images/linux/linux_09569_output_20240912_152719_original_screenshot.png
|
<image>
Click on "jupyter.enableKernelCompletions"
|
[
0.4445537328720093,
0.8042259812355042,
0.4531964957714081,
0.8193853497505188
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "jupyter.enableKernelCompletions"
|
[
958,
990,
977,
1009
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"jupyter.enableKernelCompletions\"",
"role": "user"
},
{
"content": "(967,999)",
"role": "assistant"
}
] | 2,156
| 2,156
| 1,232
|
|
doc_images/AmHHgw-Nep9dv2oIbt7q1cW53aY8_1920x1080_CHAR_19_55.png
|
<image>
Given the following text: "do you agree that “disciples are not necessarily inferior to their teachers, while teachers are not necessarily wiser than their disciples”? " , find the text in the document and click the 4th character "o" in the text.
|
[
0.5630208333333333,
0.39166666666666666,
0.5682291666666667,
0.4009259259259259
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "do you agree that “disciples are not necessarily inferior to their teachers, while teachers are not necessarily wiser than their disciples”? " , find the text in the document and click the 4th character "o" in the text.
|
[
1087,
427,
1097,
437
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"do you agree that “disciples are not necessarily inferior to their teachers, while teachers are not necessarily wiser than their disciples”? \" , find the text in the document and click the 4th character \"o\" in the text.",
"role": "user"
},
{
"content": "(1092,432)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
slides_1920*1080/slide_31/original_bbox_2.png
|
<image>
Select the rotate handle of the text box that contains the text "Maintaining composure during t...."
|
[
0.6645962732919255,
0.3882783882783883,
0.6697722567287785,
0.3974358974358974
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select the rotate handle of the text box that contains the text "Maintaining composure during t...."
|
[
1284,
424,
1294,
434
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect the rotate handle of the text box that contains the text \"Maintaining composure during t....\"",
"role": "user"
},
{
"content": "(1289,429)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
slides_3840*2160/slide_22/original.png
|
<image>
Select the handle located in the top-right corner of the text box labeled "Next Review Date: July 20XX."
|
[
0.4917184265010352,
0.28289473684210525,
0.4968944099378882,
0.29229323308270677
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select the handle located in the top-right corner of the text box labeled "Next Review Date: July 20XX."
|
[
1307,
419,
1321,
433
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect the handle located in the top-right corner of the text box labeled \"Next Review Date: July 20XX.\"",
"role": "user"
},
{
"content": "(1314,426)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
doc_images/AmHHgw-Nep9dwAhKjBwcL7W5s0_p_3840x2160_CHAR_23_3.png
|
<image>
Given the following text: "Shangda Wu," , find the text in the document and click the 1th character "n" in the text.
|
[
0.22421875,
0.23657407407407408,
0.22682291666666668,
0.2412037037037037
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "Shangda Wu," , find the text in the document and click the 1th character "n" in the text.
|
[
596,
351,
603,
357
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"Shangda Wu,\" , find the text in the document and click the 1th character \"n\" in the text.",
"role": "user"
},
{
"content": "(599,354)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
pixmo-points-images/pixmo_029743_7474b93c.png
|
<image>
Click on information button
|
[
0.33378538489341736,
0.4451163709163666,
0.3771146535873413,
0.4929272532463074
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on information button
|
[
214,
236,
242,
262
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on information button",
"role": "user"
},
{
"content": "(228,249)",
"role": "assistant"
}
] | 644
| 644
| 532
|
|
slides_1280*720/slide_72/original_bbox_0.png
|
<image>
Select the rotate handle of the text box displaying the content "MARKET EXPANSION."
|
[
0.4666149068322981,
0.28708791208791207,
0.47437888198757766,
0.3008241758241758
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select the rotate handle of the text box displaying the content "MARKET EXPANSION."
|
[
601,
208,
611,
219
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect the rotate handle of the text box displaying the content \"MARKET EXPANSION.\"",
"role": "user"
},
{
"content": "(606,213)",
"role": "assistant"
}
] | 1,280
| 1,288
| 728
|
|
autogui-images/autogui_2490031_ca84519e.png
|
<image>
Given a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element triggers the subscription process for the 'Next Pro' service, offering users the opportunity to gain increased exposure for their uploads.
|
[
0.7424761056900024,
0.09183879941701889,
0.8094406127929688,
0.12325021624565125
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element triggers the subscription process for the 'Next Pro' service, offering users the opportunity to gain increased exposure for their uploads.
|
[
623,
43,
679,
58
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element triggers the subscription process for the 'Next Pro' service, offering users the opportunity to gain increased exposure for their uploads.",
"role": "user"
},
{
"content": "(651,50)",
"role": "assistant"
}
] | 840
| 840
| 476
|
|
showui-web-images/showui_web_019377.jpg
|
<image>
Tap on Buying Format
|
[
0.7572586536407471,
0.3371105194091797,
0.8284786343574524,
0.3705913722515106
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Tap on Buying Format
|
[
2014,
500,
2203,
549
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nTap on Buying Format",
"role": "user"
},
{
"content": "(2108,524)",
"role": "assistant"
}
] | 2,796
| 2,660
| 1,484
|
|
pc-agent-e-images/data/events/screenshot/1a3b_d5f61132_11.png
|
<image>
Click the Permissions button
|
[
0.6012077927589417,
0.8894767165184021,
0.6560829877853394,
0.9272137880325317
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click the Permissions button
|
[
673,
547,
734,
571
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick the Permissions button",
"role": "user"
},
{
"content": "(703,559)",
"role": "assistant"
}
] | 1,120
| 1,120
| 616
|
|
showui-web-images/showui_web_005182.jpg
|
<image>
Tap on World
|
[
0.27937620878219604,
0.06566990166902542,
0.3146124482154846,
0.10239993780851364
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Tap on World
|
[
743,
97,
836,
151
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nTap on World",
"role": "user"
},
{
"content": "(789,124)",
"role": "assistant"
}
] | 2,717
| 2,660
| 1,484
|
|
professional-apps-grounding-images/prof_apps_0000521_frame_039273.jpg
|
<image>
Select how the photo fits within the editing window
|
[
0.6526781320571899,
0.9544405937194824,
0.7071805000305176,
1
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select how the photo fits within the editing window
|
[
1553,
1282,
1683,
1344
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect how the photo fits within the editing window",
"role": "user"
},
{
"content": "(1618,1313)",
"role": "assistant"
}
] | 2,380
| 2,380
| 1,344
|
|
doc_images/AmHHgw-Nep9dvygzDUS36KVRypAA_3840x2160_SPACE_396_7.png
|
<image>
Given the following text: "air ere " , find the text in the document and click the space between the continuous character "r" and "e" in the text.
|
[
0.23932291666666666,
0.28379629629629627,
0.24192708333333332,
0.2884259259259259
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "air ere " , find the text in the document and click the space between the continuous character "r" and "e" in the text.
|
[
636,
421,
643,
428
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"air ere \" , find the text in the document and click the space between the continuous character \"r\" and \"e\" in the text.",
"role": "user"
},
{
"content": "(639,424)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
showui-web-images/showui_web_005182.jpg
|
<image>
Choose About CNN CNN Newsletters
|
[
0.7506198287010193,
0.5589531064033508,
0.8108702898025513,
0.5747995972633362
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Choose About CNN CNN Newsletters
|
[
1996,
829,
2156,
853
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nChoose About CNN CNN Newsletters",
"role": "user"
},
{
"content": "(2076,841)",
"role": "assistant"
}
] | 2,717
| 2,660
| 1,484
|
|
osatlas-images/linux/linux_03819_output_20240912_143810_original_screenshot_sub1.png
|
<image>
Click on "Text direction from top to bottom"
|
[
0.05374865233898163,
0.17982666194438934,
0.07383489608764648,
0.22985917329788208
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "Text direction from top to bottom"
|
[
82,
156,
113,
199
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"Text direction from top to bottom\"",
"role": "user"
},
{
"content": "(97,177)",
"role": "assistant"
}
] | 1,540
| 1,540
| 868
|
|
doc_images/AmHHgw-Nep9dv1yGmW0CO81VXmM4_1280x720_CHAR_57_10.png
|
<image>
Given the following text: "Group presentation and discussion" , find the text in the document and click the 2th character "e" in the text.
|
[
0.66875,
0.7027777777777777,
0.6765625,
0.7166666666666667
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "Group presentation and discussion" , find the text in the document and click the 2th character "e" in the text.
|
[
861,
511,
871,
521
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"Group presentation and discussion\" , find the text in the document and click the 2th character \"e\" in the text.",
"role": "user"
},
{
"content": "(866,516)",
"role": "assistant"
}
] | 1,280
| 1,288
| 728
|
|
autogui-images/autogui_1690089_f0981b3d.png
|
<image>
I'll give textual clues about an element in this webpage screenshot; identify their coordinates (with point). This element acts as a navigation link, directing users to a page showcasing available extensions for the Firefox browser when clicked.
|
[
0.17933540046215057,
0.13331811130046844,
0.4497925937175751,
0.16882047057151794
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
I'll give textual clues about an element in this webpage screenshot; identify their coordinates (with point). This element acts as a navigation link, directing users to a page showcasing available extensions for the Firefox browser when clicked.
|
[
40,
48,
100,
61
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nI'll give textual clues about an element in this webpage screenshot; identify their coordinates (with point). This element acts as a navigation link, directing users to a page showcasing available extensions for the Firefox browser when clicked.",
"role": "user"
},
{
"content": "(70,54)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
autogui-images/autogui_2660579_15ea9355.png
|
<image>
Using the attached screenshot of a webpage, please find the coordinates of the described element (with point). This element provides access to information about the organizational leadership of Mozilla.
|
[
0.051303524523973465,
0.3235430121421814,
0.2629499137401581,
0.3798791468143463
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Using the attached screenshot of a webpage, please find the coordinates of the described element (with point). This element provides access to information about the organizational leadership of Mozilla.
|
[
28,
99,
147,
117
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nUsing the attached screenshot of a webpage, please find the coordinates of the described element (with point). This element provides access to information about the organizational leadership of Mozilla.",
"role": "user"
},
{
"content": "(87,108)",
"role": "assistant"
}
] | 560
| 560
| 308
|
|
slides_1920*1080/slide_15/original.png
|
<image>
Select the handle located at the bottom-left corner of the text box labeled "Presentation title."
|
[
0.2515527950310559,
0.2591575091575092,
0.2567287784679089,
0.2683150183150183
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select the handle located at the bottom-left corner of the text box labeled "Presentation title."
|
[
485,
283,
496,
293
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect the handle located at the bottom-left corner of the text box labeled \"Presentation title.\"",
"role": "user"
},
{
"content": "(490,288)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
slides_1280*720/slide_56/original.png
|
<image>
Select the handle located at the bottom of the text box that contains the text "Marketing strategies."
|
[
0.7368012422360248,
0.4574175824175824,
0.7445652173913043,
0.47115384615384615
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select the handle located at the bottom of the text box that contains the text "Marketing strategies."
|
[
949,
333,
959,
343
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect the handle located at the bottom of the text box that contains the text \"Marketing strategies.\"",
"role": "user"
},
{
"content": "(954,338)",
"role": "assistant"
}
] | 1,280
| 1,288
| 728
|
|
waveui-images/waveui_011814.jpg
|
<image>
Select to indicate that the field is for entering passwords
|
[
0.7239466905593872,
0.5153990387916565,
0.7543578743934631,
0.5660013556480408
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select to indicate that the field is for entering passwords
|
[
425,
230,
443,
253
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect to indicate that the field is for entering passwords",
"role": "user"
},
{
"content": "(434,241)",
"role": "assistant"
}
] | 579
| 588
| 448
|
|
omniact-images/desktop/reminder/screen_1.png
|
<image>
Add the event for tomorrow.
|
[
0.20303116738796234,
0.44256654381752014,
0.22331203520298004,
0.4807065427303314
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Add the event for tomorrow.
|
[
511,
693,
562,
753
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nAdd the event for tomorrow.",
"role": "user"
},
{
"content": "(536,723)",
"role": "assistant"
}
] | 2,884
| 2,520
| 1,568
|
|
autogui-images/autogui_051152_6cdac501.png
|
<image>
I'll give textual clues about an element in this webpage screenshot; identify their coordinates (with point). This element serves as a filter button for users to apply a specific filter to the search results. By clicking on it, users can view all the search results, regardless of the filter type. This functionality helps users to have an overview of all the search results and then narrow down their search using other filters if needed.
|
[
0.06957283616065979,
0.36882948875427246,
0.14945900440216064,
0.43813738226890564
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
I'll give textual clues about an element in this webpage screenshot; identify their coordinates (with point). This element serves as a filter button for users to apply a specific filter to the search results. By clicking on it, users can view all the search results, regardless of the filter type. This functionality helps users to have an overview of all the search results and then narrow down their search using other filters if needed.
|
[
15,
134,
33,
159
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nI'll give textual clues about an element in this webpage screenshot; identify their coordinates (with point). This element serves as a filter button for users to apply a specific filter to the search results. By clicking on it, users can view all the search results, regardless of the filter type. This functionality helps users to have an overview of all the search results and then narrow down their search using other filters if needed.",
"role": "user"
},
{
"content": "(24,146)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
autogui-images/autogui_1811213_61209d62.png
|
<image>
Given a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element acts as a navigation tool, directing users to the TikTok platform when clicked, thereby facilitating easy access to the main TikTok website or application.
|
[
0.09198515117168427,
0.006984639447182417,
0.43881964683532715,
0.08308234810829163
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element acts as a navigation tool, directing users to the TikTok platform when clicked, thereby facilitating easy access to the main TikTok website or application.
|
[
20,
2,
98,
30
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element acts as a navigation tool, directing users to the TikTok platform when clicked, thereby facilitating easy access to the main TikTok website or application.",
"role": "user"
},
{
"content": "(59,16)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
osatlas-images/linux/linux_00145_output_20240912_152955_original_screenshot_sub1.png
|
<image>
Click on "More options menu"
|
[
0.8220466375350952,
0.17259567975997925,
0.8527427315711975,
0.23546023666858673
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "More options menu"
|
[
782,
91,
811,
125
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"More options menu\"",
"role": "user"
},
{
"content": "(796,108)",
"role": "assistant"
}
] | 952
| 952
| 532
|
|
uground-images/uground_112018.jpg
|
<image>
Tap on Sports
|
[
0.7346341609954834,
0.31309574842453003,
0.9647129774093628,
0.34543928503990173
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Tap on Sports
|
[
843,
561,
1107,
619
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nTap on Sports",
"role": "user"
},
{
"content": "(975,590)",
"role": "assistant"
}
] | 1,151
| 1,148
| 1,792
|
|
cell_left_edge/images/1920x1080_cell_B29.png
|
<image>
Left boundary of cell B29
|
[
0.06418219461697723,
0.6932234432234432,
0.06935817805383022,
0.7023809523809523
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Left boundary of cell B29
|
[
124,
757,
134,
766
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nLeft boundary of cell B29",
"role": "user"
},
{
"content": "(129,761)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
pc-agent-e-images/data/events/screenshot/9d83_8377f75b_7.png
|
<image>
Click on 辅助功能 (Accessibility)
|
[
0.0036693094298243523,
0.6118313670158386,
0.26577332615852356,
0.6584903001785278
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on 辅助功能 (Accessibility)
|
[
2,
274,
208,
295
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on 辅助功能 (Accessibility)",
"role": "user"
},
{
"content": "(105,284)",
"role": "assistant"
}
] | 784
| 784
| 448
|
|
showui-desktop-images/showui_desktop_002595.jpg
|
<image>
Point to Get driving directions to The Regent's Park.
|
[
0.8587595224380493,
0.06997954100370407,
0.8876012563705444,
0.10888385772705078
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Point to Get driving directions to The Regent's Park.
|
[
2019,
101,
2087,
158
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nPoint to Get driving directions to The Regent's Park.",
"role": "user"
},
{
"content": "(2053,129)",
"role": "assistant"
}
] | 2,342
| 2,352
| 1,456
|
|
autogui-images/autogui_2222014_19d2e4aa.png
|
<image>
Using the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element initiates a search on YouTube when interacted with, providing users a means to discover and access various video content on the platform.
|
[
0.7744863629341125,
0.0013556837802752852,
0.8855946660041809,
0.07289253920316696
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Using the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element initiates a search on YouTube when interacted with, providing users a means to discover and access various video content on the platform.
|
[
173,
0,
198,
26
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nUsing the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element initiates a search on YouTube when interacted with, providing users a means to discover and access various video content on the platform.",
"role": "user"
},
{
"content": "(185,13)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
professional-apps-grounding-images/prof_apps_0000791_frame_004845.jpg
|
<image>
Resume playing from the current playhead position
|
[
0.5487076640129089,
0.5441470742225647,
0.5615349411964417,
0.569193959236145
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Resume playing from the current playhead position
|
[
1121,
624,
1147,
653
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nResume playing from the current playhead position",
"role": "user"
},
{
"content": "(1134,638)",
"role": "assistant"
}
] | 2,044
| 2,044
| 1,148
|
|
showui-web-images/showui_web_005443.jpg
|
<image>
Choose Listen
|
[
0.7200771570205688,
0.06659044325351715,
0.7540378570556641,
0.09982918202877045
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Choose Listen
|
[
1794,
93,
1879,
139
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nChoose Listen",
"role": "user"
},
{
"content": "(1836,116)",
"role": "assistant"
}
] | 2,482
| 2,492
| 1,400
|
|
doc_images/AmHHgw-Nep9dv1nz-z2lix187ekR_1920x1080_CHAR_29_19.png
|
<image>
Given the following text: " decide too early that you dislike what you are reading. The first few pages will seldom have the clarity and impact of a short story. Read attentively, with an open mind. Refrain from insisting that the author " , find the text in the document and click the 1th character "h" in the text.
|
[
0.5979166666666667,
0.5925925925925926,
0.603125,
0.6018518518518519
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: " decide too early that you dislike what you are reading. The first few pages will seldom have the clarity and impact of a short story. Read attentively, with an open mind. Refrain from insisting that the author " , find the text in the document and click the 1th character "h" in the text.
|
[
1155,
647,
1165,
657
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \" decide too early that you dislike what you are reading. The first few pages will seldom have the clarity and impact of a short story. Read attentively, with an open mind. Refrain from insisting that the author \" , find the text in the document and click the 1th character \"h\" in the text.",
"role": "user"
},
{
"content": "(1160,652)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
slides_1280*720/slide_11/original.png
|
<image>
Drag the handle located at the bottom-right corner of the text box containing the text "With this product, Adatum Corp...."
|
[
0.7057453416149069,
0.7802197802197802,
0.7135093167701864,
0.7939560439560439
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Drag the handle located at the bottom-right corner of the text box containing the text "With this product, Adatum Corp...."
|
[
909,
568,
919,
578
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nDrag the handle located at the bottom-right corner of the text box containing the text \"With this product, Adatum Corp....\"",
"role": "user"
},
{
"content": "(914,573)",
"role": "assistant"
}
] | 1,280
| 1,288
| 728
|
|
seeclick-images/58ed66283dfc62a14fcf7c907d8eed52.png
|
<image>
Click on "Cart
|
[
0.8001458644866943,
0.04729513078927994,
0.8290311694145203,
0.13865597546100616
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "Cart
|
[
1837,
60,
1903,
178
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"Cart",
"role": "user"
},
{
"content": "(1870,119)",
"role": "assistant"
}
] | 2,296
| 2,296
| 1,288
|
|
uground-images/uground_196795.jpg
|
<image>
Press Simple black heart icon on white background, centered and modern
|
[
0.8177664279937744,
0.04466697946190834,
0.8691065907478333,
0.07153170555830002
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Press Simple black heart icon on white background, centered and modern
|
[
686,
77,
730,
124
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nPress Simple black heart icon on white background, centered and modern",
"role": "user"
},
{
"content": "(708,100)",
"role": "assistant"
}
] | 832
| 840
| 1,736
|
|
uground-images/uground_639122.jpg
|
<image>
Identify forreachingme is offline, below "birding"
|
[
0.735234797000885,
0.49427786469459534,
0.7983869314193726,
0.5104766488075256
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Identify forreachingme is offline, below "birding"
|
[
988,
484,
1073,
500
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nIdentify forreachingme is offline, below \"birding\"",
"role": "user"
},
{
"content": "(1030,492)",
"role": "assistant"
}
] | 1,330
| 1,344
| 980
|
|
uground-images/uground_700737.jpg
|
<image>
Tap on button for "MCKESSON"
|
[
0.5328624248504639,
0.2878687083721161,
0.5916144847869873,
0.3294213116168976
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Tap on button for "MCKESSON"
|
[
969,
354,
1076,
405
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nTap on button for \"MCKESSON\"",
"role": "user"
},
{
"content": "(1022,379)",
"role": "assistant"
}
] | 1,825
| 1,820
| 1,232
|
|
osatlas-images/linux/linux_02647_output_20240912_113714_original_screenshot.png
|
<image>
Click on "Font Color"
|
[
0.3124725818634033,
0.08942652493715286,
0.32665470242500305,
0.11204327642917633
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "Font Color"
|
[
831,
132,
868,
166
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"Font Color\"",
"role": "user"
},
{
"content": "(849,149)",
"role": "assistant"
}
] | 3,332
| 2,660
| 1,484
|
|
osatlas-images/linux/linux_04824_output_20240912_145808_original_screenshot_sub0.png
|
<image>
Click on "Fontwork Style"
|
[
0.627590537071228,
0.12808972597122192,
0.6511523723602295,
0.1689528524875641
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "Fontwork Style"
|
[
1089,
125,
1130,
165
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"Fontwork Style\"",
"role": "user"
},
{
"content": "(1109,145)",
"role": "assistant"
}
] | 1,736
| 1,736
| 980
|
|
slides_1920*1080/slide_32/original.png
|
<image>
Select the handle located at the top of the text box containing the text "Your ability to communicate ef...".
|
[
0.432712215320911,
0.40201465201465203,
0.43788819875776397,
0.41117216117216115
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select the handle located at the top of the text box containing the text "Your ability to communicate ef...".
|
[
836,
439,
846,
449
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect the handle located at the top of the text box containing the text \"Your ability to communicate ef...\".",
"role": "user"
},
{
"content": "(841,444)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
autogui-images/autogui_1361256_20896203.png
|
<image>
Based on the screenshot of the page, I give a text description and you give its corresponding location (with point). This element provides access to a list of events related to the EU energy market.
|
[
0.04387017339468002,
0.13123324513435364,
0.12044195830821991,
0.1722572147846222
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Based on the screenshot of the page, I give a text description and you give its corresponding location (with point). This element provides access to a list of events related to the EU energy market.
|
[
20,
36,
57,
48
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nBased on the screenshot of the page, I give a text description and you give its corresponding location (with point). This element provides access to a list of events related to the EU energy market.",
"role": "user"
},
{
"content": "(38,42)",
"role": "assistant"
}
] | 476
| 476
| 280
|
|
waveui-images/waveui_005195.jpg
|
<image>
Select to provide additional information
|
[
0.3056367337703705,
0.00021583636407740414,
0.4528103768825531,
0.03629475459456444
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select to provide additional information
|
[
445,
0,
659,
39
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect to provide additional information",
"role": "user"
},
{
"content": "(552,19)",
"role": "assistant"
}
] | 1,460
| 1,456
| 1,092
|
|
uground-images/uground_152340.jpg
|
<image>
Identify the image of "White arrow button with gradient fill, set against light blue background", on the left side of "Is Standard Bank Autoplus transaction a reliable proof of payment"
|
[
0.00404107291251421,
0.093671053647995,
0.03338614106178284,
0.11152035742998123
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Identify the image of "White arrow button with gradient fill, set against light blue background", on the left side of "Is Standard Bank Autoplus transaction a reliable proof of payment"
|
[
1,
70,
14,
84
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nIdentify the image of \"White arrow button with gradient fill, set against light blue background\", on the left side of \"Is Standard Bank Autoplus transaction a reliable proof of payment\"",
"role": "user"
},
{
"content": "(7,77)",
"role": "assistant"
}
] | 409
| 420
| 756
|
|
autogui-images/autogui_1751446_83426d01.png
|
<image>
On this webpage screenshot, I will point out an element; please predict their exact coordinates (with point). This element provides access to a document outlining the platform's privacy practices, allowing users to understand how their personal data is handled.
|
[
0.01589066907763481,
0.5089771747589111,
0.2774621844291687,
0.5858088135719299
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
On this webpage screenshot, I will point out an element; please predict their exact coordinates (with point). This element provides access to a document outlining the platform's privacy practices, allowing users to understand how their personal data is handled.
|
[
6,
114,
108,
131
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nOn this webpage screenshot, I will point out an element; please predict their exact coordinates (with point). This element provides access to a document outlining the platform's privacy practices, allowing users to understand how their personal data is handled.",
"role": "user"
},
{
"content": "(57,122)",
"role": "assistant"
}
] | 392
| 392
| 224
|
|
pixmo-points-images/pixmo_1303638_abd4e0d5.png
|
<image>
Click on free shipping hyperlink
|
[
0.7938319444656372,
0.8400343656539917,
0.8827522993087769,
0.8624376654624939
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on free shipping hyperlink
|
[
1244,
1081,
1384,
1110
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on free shipping hyperlink",
"role": "user"
},
{
"content": "(1314,1095)",
"role": "assistant"
}
] | 1,568
| 1,568
| 1,288
|
|
pc-agent-e-images/data/events/screenshot/f31f_33109747_3.png
|
<image>
Turn on Wi-Fi toggle switch
|
[
0.9537944197654724,
0.37361061573028564,
0.9839015603065491,
0.4182368218898773
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Turn on Wi-Fi toggle switch
|
[
1308,
282,
1349,
316
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nTurn on Wi-Fi toggle switch",
"role": "user"
},
{
"content": "(1328,299)",
"role": "assistant"
}
] | 1,372
| 1,372
| 756
|
|
doc_images/AmHHgw-Nep9dv0zI5NGLXbhP8S_E_1920x1080_CHAR_94_34.png
|
<image>
Given the following text: " to consider in relation to chocolate is the " , find the text in the document and click the 2th character "a" in the text.
|
[
0.54375,
0.725925925925926,
0.5489583333333333,
0.7351851851851852
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: " to consider in relation to chocolate is the " , find the text in the document and click the 2th character "a" in the text.
|
[
1050,
792,
1060,
802
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \" to consider in relation to chocolate is the \" , find the text in the document and click the 2th character \"a\" in the text.",
"role": "user"
},
{
"content": "(1055,797)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
autogui-images/autogui_2602457_1d2d8420.png
|
<image>
In this web page image, please locate the element as I describe it (with point). This element serves as a navigation tool, guiding users to the product page for the 'airMAX 2.4 GHz' product. Interacting with this link would lead the user to a new webpage dedicated to the details, specifications, and potential purchase options for the 'airMAX 2.4 GHz' product.
|
[
0.1299522966146469,
0.06284859776496887,
0.4476853907108307,
0.17538100481033325
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
In this web page image, please locate the element as I describe it (with point). This element serves as a navigation tool, guiding users to the product page for the 'airMAX 2.4 GHz' product. Interacting with this link would lead the user to a new webpage dedicated to the details, specifications, and potential purchase options for the 'airMAX 2.4 GHz' product.
|
[
29,
22,
100,
63
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nIn this web page image, please locate the element as I describe it (with point). This element serves as a navigation tool, guiding users to the product page for the 'airMAX 2.4 GHz' product. Interacting with this link would lead the user to a new webpage dedicated to the details, specifications, and potential purchase options for the 'airMAX 2.4 GHz' product.",
"role": "user"
},
{
"content": "(64,42)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
pixmo-points-images/pixmo_1039225_ce6c2628.png
|
<image>
Click on left cloud icon
|
[
0.005881261546164751,
0.34625720977783203,
0.47482696175575256,
0.8908081650733948
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on left cloud icon
|
[
3,
96,
292,
249
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on left cloud icon",
"role": "user"
},
{
"content": "(147,172)",
"role": "assistant"
}
] | 616
| 616
| 280
|
|
cell_right_edge/images/1920x1080_cell_M58.png
|
<image>
Select right border of M58
|
[
0.5641821946169773,
0.8873626373626373,
0.5693581780538303,
0.8965201465201466
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select right border of M58
|
[
1090,
969,
1100,
979
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect right border of M58",
"role": "user"
},
{
"content": "(1095,974)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
omniact-images/desktop/system_preferences/screen_1.png
|
<image>
Proceed to the subsequent page in system preferences.
|
[
0.3088000416755676,
0.08399777859449387,
0.3350306451320648,
0.12926624715328217
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Proceed to the subsequent page in system preferences.
|
[
778,
131,
844,
202
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nProceed to the subsequent page in system preferences.",
"role": "user"
},
{
"content": "(811,166)",
"role": "assistant"
}
] | 2,884
| 2,520
| 1,568
|
|
showui-web-images/showui_web_009510.jpg
|
<image>
Find and click Career Insights & Advice Hub
|
[
0.6785473227500916,
0.12160835415124893,
0.7867923974990845,
0.150030717253685
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Find and click Career Insights & Advice Hub
|
[
664,
68,
771,
84
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nFind and click Career Insights & Advice Hub",
"role": "user"
},
{
"content": "(717,76)",
"role": "assistant"
}
] | 976
| 980
| 560
|
|
cell_right_bottom_corner/images/1920x1080_cell_D69.png
|
<image>
Southeast corner of cell D69
|
[
0.18944099378881987,
0.826007326007326,
0.19461697722567287,
0.8351648351648352
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Southeast corner of cell D69
|
[
366,
902,
376,
912
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSoutheast corner of cell D69",
"role": "user"
},
{
"content": "(371,907)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
pc-agent-e-images/data/events/screenshot/d41a_93cd75dd_7.png
|
<image>
Check the Sepia checkbox
|
[
0.2681524455547333,
0.45433083176612854,
0.2776438891887665,
0.4722461402416229
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Check the Sepia checkbox
|
[
465,
445,
481,
462
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nCheck the Sepia checkbox",
"role": "user"
},
{
"content": "(473,453)",
"role": "assistant"
}
] | 1,736
| 1,736
| 980
|
|
uground-images/uground_106600.jpg
|
<image>
Navigate to Click Must Read, on the left side of the screenshot
|
[
0.03363346308469772,
0.637282133102417,
0.17673398554325104,
0.6488857269287109
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Navigate to Click Must Read, on the left side of the screenshot
|
[
16,
945,
89,
962
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nNavigate to Click Must Read, on the left side of the screenshot",
"role": "user"
},
{
"content": "(52,953)",
"role": "assistant"
}
] | 516
| 504
| 1,484
|
|
autogui-images/autogui_1081386_bd369d16.png
|
<image>
I will describe an element from a webpage screenshot; your role is to locate it (with point). This element sets the AI assistant's conversation style to a balanced approach, aiming for a tone that strikes a middle ground between creativity and precision in responses.
|
[
0.39950230717658997,
0.6410340070724487,
0.6117076277732849,
0.7213260531425476
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
I will describe an element from a webpage screenshot; your role is to locate it (with point). This element sets the AI assistant's conversation style to a balanced approach, aiming for a tone that strikes a middle ground between creativity and precision in responses.
|
[
89,
233,
137,
262
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nI will describe an element from a webpage screenshot; your role is to locate it (with point). This element sets the AI assistant's conversation style to a balanced approach, aiming for a tone that strikes a middle ground between creativity and precision in responses.",
"role": "user"
},
{
"content": "(113,247)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
pc-agent-e-images/data/events/screenshot/b102_34da7940_1.png
|
<image>
Click the Chrome menu button
|
[
0.9783673882484436,
0.051449816673994064,
0.9948709607124329,
0.08718832582235336
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click the Chrome menu button
|
[
1643,
48,
1671,
83
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick the Chrome menu button",
"role": "user"
},
{
"content": "(1657,65)",
"role": "assistant"
}
] | 1,680
| 1,680
| 952
|
|
doc_images/AmHHgw-Nep9dv2Zzal4Im081GzBe_3840x2160_SPACE_110_20.png
|
<image>
Given the following text: " to be handwritten using " , find the text in the document and click the space between the continuous character "s" and "i" in the text.
|
[
0.26640625,
0.2962962962962963,
0.26901041666666664,
0.30092592592592593
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: " to be handwritten using " , find the text in the document and click the space between the continuous character "s" and "i" in the text.
|
[
708,
439,
715,
446
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \" to be handwritten using \" , find the text in the document and click the space between the continuous character \"s\" and \"i\" in the text.",
"role": "user"
},
{
"content": "(711,442)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
doc_images/AmHHgw-Nep9dv2FHRpQ58IDlI0A3_1920x1080_SPACE_4_78.png
|
<image>
Given the following text: " independently, moving close to each other as the novel progresses until they become entangled halfway through or near the end." , find the text in the document and click the space between the continuous character "b" and "e" in the text.
|
[
0.4125,
0.3925925925925926,
0.41770833333333335,
0.40185185185185185
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: " independently, moving close to each other as the novel progresses until they become entangled halfway through or near the end." , find the text in the document and click the space between the continuous character "b" and "e" in the text.
|
[
796,
428,
807,
438
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \" independently, moving close to each other as the novel progresses until they become entangled halfway through or near the end.\" , find the text in the document and click the space between the continuous character \"b\" and \"e\" in the text.",
"role": "user"
},
{
"content": "(801,433)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
osatlas-images/linux/linux_01947_output_20240912_160618_original_screenshot.png
|
<image>
Click on "Search"
|
[
0.5781311988830566,
0.10430330783128738,
0.5919596552848816,
0.123753622174263
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "Search"
|
[
809,
81,
828,
97
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"Search\"",
"role": "user"
},
{
"content": "(818,89)",
"role": "assistant"
}
] | 1,400
| 1,400
| 784
|
|
showui-web-images/showui_web_012975.jpg
|
<image>
Point to AirTag
|
[
0.1739615499973297,
0.6524655818939209,
0.29327479004859924,
0.6786165237426758
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Point to AirTag
|
[
228,
493,
385,
513
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nPoint to AirTag",
"role": "user"
},
{
"content": "(306,503)",
"role": "assistant"
}
] | 1,326
| 1,316
| 756
|
|
waveui-images/waveui_001623.jpg
|
<image>
Click to navigate to the contact page
|
[
0.7538992166519165,
0.12724018096923828,
0.8012721538543701,
0.1646532267332077
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click to navigate to the contact page
|
[
971,
92,
1032,
119
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick to navigate to the contact page",
"role": "user"
},
{
"content": "(1001,105)",
"role": "assistant"
}
] | 1,297
| 1,288
| 728
|
|
showui-web-images/showui_web_003518.jpg
|
<image>
Locate Privacy
|
[
0.3114537000656128,
0.8426210880279541,
0.33766117691993713,
0.8642127513885498
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Locate Privacy
|
[
697,
1061,
756,
1088
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nLocate Privacy",
"role": "user"
},
{
"content": "(726,1074)",
"role": "assistant"
}
] | 2,234
| 2,240
| 1,260
|
|
uground-images/uground_053046.jpg
|
<image>
Point to county (15 items)
|
[
0.6317594051361084,
0.3614736795425415,
0.707284152507782,
0.3828938603401184
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Point to county (15 items)
|
[
406,
455,
455,
482
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nPoint to county (15 items)",
"role": "user"
},
{
"content": "(430,468)",
"role": "assistant"
}
] | 646
| 644
| 1,260
|
|
doc_images/AmHHgw-Nep9dvymYjyXkKaOerDIy_1920x1080_CHAR_11_8.png
|
<image>
Given the following text: "Example essay" , find the text in the document and click the 2th character "e" in the text.
|
[
0.4010416666666667,
0.31574074074074077,
0.40625,
0.325
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "Example essay" , find the text in the document and click the 2th character "e" in the text.
|
[
774,
344,
784,
354
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"Example essay\" , find the text in the document and click the 2th character \"e\" in the text.",
"role": "user"
},
{
"content": "(779,349)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
slides_3840*2160/slide_67/original.png
|
<image>
Click on the handle at the top-right corner of the bounding box surrounding the image of a boat's bow cutting through calm ocean waters, with a distant shoreline and clear blue sky in the background.
|
[
0.7474120082815735,
0.606203007518797,
0.7525879917184265,
0.6156015037593985
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on the handle at the top-right corner of the bounding box surrounding the image of a boat's bow cutting through calm ocean waters, with a distant shoreline and clear blue sky in the background.
|
[
1988,
899,
2001,
913
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on the handle at the top-right corner of the bounding box surrounding the image of a boat's bow cutting through calm ocean waters, with a distant shoreline and clear blue sky in the background.",
"role": "user"
},
{
"content": "(1994,906)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
pixmo-points-images/pixmo_047694_fb940b44.png
|
<image>
Click on Green checkmarks icon
|
[
0.5042508244514465,
0.9243452548980713,
0.5368509888648987,
0.9687815308570862
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on Green checkmarks icon
|
[
917,
1087,
977,
1139
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on Green checkmarks icon",
"role": "user"
},
{
"content": "(947,1113)",
"role": "assistant"
}
] | 1,820
| 1,820
| 1,176
|
|
autogui-images/autogui_940812_a2782ba0.png
|
<image>
On this webpage screenshot, I will point out an element; please predict their exact coordinates (with point). This element provides access to information about Firefox browsers on mobile devices.
|
[
0.04491296410560608,
0.1090017557144165,
0.16120679676532745,
0.14462962746620178
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
On this webpage screenshot, I will point out an element; please predict their exact coordinates (with point). This element provides access to information about Firefox browsers on mobile devices.
|
[
35,
48,
126,
64
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nOn this webpage screenshot, I will point out an element; please predict their exact coordinates (with point). This element provides access to information about Firefox browsers on mobile devices.",
"role": "user"
},
{
"content": "(80,56)",
"role": "assistant"
}
] | 784
| 784
| 448
|
|
showui-web-images/showui_web_006716.jpg
|
<image>
Navigate to Lyon
|
[
0.5439621210098267,
0.8814646005630493,
0.5656970739364624,
0.9023441076278687
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Navigate to Lyon
|
[
670,
617,
696,
631
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nNavigate to Lyon",
"role": "user"
},
{
"content": "(683,624)",
"role": "assistant"
}
] | 1,232
| 1,232
| 700
|
|
doc_images/AmHHgw-Nep9dv1nz-z2lix187ekR_3840x2160_CHAR_110_5.png
|
<image>
Given the following text: "What happens when I die?" , find the text in the document and click the 2th character "h" in the text.
|
[
0.21458333333333332,
0.42268518518518516,
0.2171875,
0.4273148148148148
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "What happens when I die?" , find the text in the document and click the 2th character "h" in the text.
|
[
570,
627,
577,
634
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"What happens when I die?\" , find the text in the document and click the 2th character \"h\" in the text.",
"role": "user"
},
{
"content": "(573,630)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
autogui-images/autogui_1341756_1fada269.png
|
<image>
In this web page image, please locate the element as I describe it (with point). This element allows users to choose a language preference for the webpage, providing a personalized user experience.
|
[
0.39022815227508545,
0.022591199725866318,
0.4197580814361572,
0.07803338021039963
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
In this web page image, please locate the element as I describe it (with point). This element allows users to choose a language preference for the webpage, providing a personalized user experience.
|
[
218,
6,
235,
24
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nIn this web page image, please locate the element as I describe it (with point). This element allows users to choose a language preference for the webpage, providing a personalized user experience.",
"role": "user"
},
{
"content": "(226,15)",
"role": "assistant"
}
] | 560
| 560
| 308
|
|
doc_images/AmHHgw-Nep9dv1H8Pl8chRcKjDOa_1280x720_CHAR_15_97.png
|
<image>
Given the following text: "With governments today seeking effective ways to lead their nations out of these related crises whilst also " , find the text in the document and click the 4th character "h" in the text.
|
[
0.2953125,
0.8055555555555556,
0.303125,
0.8194444444444444
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "With governments today seeking effective ways to lead their nations out of these related crises whilst also " , find the text in the document and click the 4th character "h" in the text.
|
[
380,
586,
390,
596
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"With governments today seeking effective ways to lead their nations out of these related crises whilst also \" , find the text in the document and click the 4th character \"h\" in the text.",
"role": "user"
},
{
"content": "(385,591)",
"role": "assistant"
}
] | 1,280
| 1,288
| 728
|
|
cell_left_top_corner/images/1920x1080_cell_V5.png
|
<image>
Top left corner of V-5
|
[
0.8975155279503105,
0.23626373626373626,
0.9026915113871635,
0.2454212454212454
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Top left corner of V-5
|
[
1734,
258,
1744,
268
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nTop left corner of V-5",
"role": "user"
},
{
"content": "(1739,263)",
"role": "assistant"
}
] | 1,920
| 1,932
| 1,092
|
|
slides_3840*2160/slide_67/original.png
|
<image>
Click on the handle on the bottom side of the bounding box surrounding the photograph of a person sitting on a rocky cliff overlooking the ocean during daylight.
|
[
0.5926501035196687,
0.756578947368421,
0.5978260869565217,
0.7659774436090225
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on the handle on the bottom side of the bounding box surrounding the photograph of a person sitting on a rocky cliff overlooking the ocean during daylight.
|
[
1576,
1122,
1590,
1136
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on the handle on the bottom side of the bounding box surrounding the photograph of a person sitting on a rocky cliff overlooking the ocean during daylight.",
"role": "user"
},
{
"content": "(1583,1129)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
autogui-images/autogui_131595_3065b1c7.png
|
<image>
Given a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element serves as a navigation tool for users to access their shopping cart or wishlist, providing a quick overview of their current item selection.
|
[
0.7892447710037231,
0.010707736946642399,
0.8883156180381775,
0.06262648850679398
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element serves as a navigation tool for users to access their shopping cart or wishlist, providing a quick overview of their current item selection.
|
[
176,
3,
198,
22
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven a screenshot, I will describe a specific element; your task is to predict their locations (with point). This element serves as a navigation tool for users to access their shopping cart or wishlist, providing a quick overview of their current item selection.",
"role": "user"
},
{
"content": "(187,12)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
showui-web-images/showui_web_008011.jpg
|
<image>
Tap on 13 hours • Pickup available • Optional audio guide
|
[
0.5048564076423645,
0.6087808012962341,
0.650442361831665,
0.6523605585098267
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Tap on 13 hours • Pickup available • Optional audio guide
|
[
565,
375,
728,
401
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nTap on 13 hours • Pickup available • Optional audio guide",
"role": "user"
},
{
"content": "(646,388)",
"role": "assistant"
}
] | 1,120
| 1,120
| 616
|
|
autogui-images/autogui_432415_f24d9281.png
|
<image>
In this web page capture, please predict the positions of the element I describe (with point). This element serves as a navigation link to the 'DelmarvaLife' section, which likely provides content related to the lifestyle of the Delmarva region. Interacting with this link would direct users to the 'DelmarvaLife' section of the website, offering them access to related articles, features, or other content.
|
[
0.32871201634407043,
0.25243261456489563,
0.4214106500148773,
0.3244304358959198
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
In this web page capture, please predict the positions of the element I describe (with point). This element serves as a navigation link to the 'DelmarvaLife' section, which likely provides content related to the lifestyle of the Delmarva region. Interacting with this link would direct users to the 'DelmarvaLife' section of the website, offering them access to related articles, features, or other content.
|
[
239,
98,
306,
127
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nIn this web page capture, please predict the positions of the element I describe (with point). This element serves as a navigation link to the 'DelmarvaLife' section, which likely provides content related to the lifestyle of the Delmarva region. Interacting with this link would direct users to the 'DelmarvaLife' section of the website, offering them access to related articles, features, or other content.",
"role": "user"
},
{
"content": "(272,112)",
"role": "assistant"
}
] | 728
| 728
| 392
|
|
omniact-images/desktop/weather/screen_2.png
|
<image>
Access the feels like temperature indicator for New Delhi
|
[
0.5032563805580139,
0.6960952281951904,
0.5925248265266418,
0.8393047451972961
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Access the feels like temperature indicator for New Delhi
|
[
1268,
1091,
1493,
1316
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nAccess the feels like temperature indicator for New Delhi",
"role": "user"
},
{
"content": "(1380,1203)",
"role": "assistant"
}
] | 3,360
| 2,520
| 1,568
|
|
autogui-images/autogui_1241027_e07dbf63.png
|
<image>
Given this screenshot of a webpage, I'll describe an element; locate it for me (with point). This element serves as a navigation link that directs users to articles and news related to the economy when clicked or selected. It allows users to access a collection of articles and news focused on the economy within the CNN Business website.
|
[
0.08482483774423599,
0.27283114194869995,
0.13741673529148102,
0.3066205680370331
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given this screenshot of a webpage, I'll describe an element; locate it for me (with point). This element serves as a navigation link that directs users to articles and news related to the economy when clicked or selected. It allows users to access a collection of articles and news focused on the economy within the CNN Business website.
|
[
71,
129,
115,
145
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven this screenshot of a webpage, I'll describe an element; locate it for me (with point). This element serves as a navigation link that directs users to articles and news related to the economy when clicked or selected. It allows users to access a collection of articles and news focused on the economy within the CNN Business website.",
"role": "user"
},
{
"content": "(93,137)",
"role": "assistant"
}
] | 840
| 840
| 476
|
|
autogui-images/autogui_1392422_c1c4168c.png
|
<image>
I will describe features on this webpage screenshot; please predict their positions (with point). This element acts as a gateway for users to sign up for a free account on the SoundCloud platform, allowing them to access and interact with the music streaming service.
|
[
0.024460483342409134,
0.5975387096405029,
0.9688674211502075,
0.6692350506782532
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
I will describe features on this webpage screenshot; please predict their positions (with point). This element acts as a gateway for users to sign up for a free account on the SoundCloud platform, allowing them to access and interact with the music streaming service.
|
[
5,
217,
217,
243
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nI will describe features on this webpage screenshot; please predict their positions (with point). This element acts as a gateway for users to sign up for a free account on the SoundCloud platform, allowing them to access and interact with the music streaming service.",
"role": "user"
},
{
"content": "(111,230)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
autogui-images/autogui_750843_0444cfe5.png
|
<image>
I'll provide textual descriptions of the element in this webpage screenshot. Can you find their coordinates (with point)? This element acts as a navigation shortcut, guiding users to a dedicated page or section showcasing Apple products on the apple.com website.
|
[
0.013378052040934563,
0.0028343976009637117,
0.10177766531705856,
0.060757122933864594
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
I'll provide textual descriptions of the element in this webpage screenshot. Can you find their coordinates (with point)? This element acts as a navigation shortcut, guiding users to a dedicated page or section showcasing Apple products on the apple.com website.
|
[
2,
1,
22,
22
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nI'll provide textual descriptions of the element in this webpage screenshot. Can you find their coordinates (with point)? This element acts as a navigation shortcut, guiding users to a dedicated page or section showcasing Apple products on the apple.com website.",
"role": "user"
},
{
"content": "(12,11)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
showui-web-images/showui_web_000763.jpg
|
<image>
Navigate to 10, Sunday, November 2024. Unavailable
|
[
0.5180515050888062,
0.4273505210876465,
0.5366491079330444,
0.46221858263015747
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Navigate to 10, Sunday, November 2024. Unavailable
|
[
1116,
514,
1157,
556
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nNavigate to 10, Sunday, November 2024. Unavailable",
"role": "user"
},
{
"content": "(1136,535)",
"role": "assistant"
}
] | 2,159
| 2,156
| 1,204
|
|
osatlas-images/linux/linux_01089_output_20240912_153836_original_screenshot_sub0.png
|
<image>
Click on "Files"
|
[
0.003046028083190322,
0.24010246992111206,
0.05126732960343361,
0.3194012939929962
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "Files"
|
[
3,
147,
55,
196
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"Files\"",
"role": "user"
},
{
"content": "(29,171)",
"role": "assistant"
}
] | 1,092
| 1,092
| 616
|
|
autogui-images/autogui_1082062_7e84f331.png
|
<image>
Using the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element acts as a navigation link, directing users to specific pages or resources upon activation, contributing to a seamless browsing experience within the Cloudflare webpage.
|
[
0.025473343208432198,
0.019703244790434837,
0.22857967019081116,
0.09284792095422745
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Using the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element acts as a navigation link, directing users to specific pages or resources upon activation, contributing to a seamless browsing experience within the Cloudflare webpage.
|
[
5,
7,
51,
33
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nUsing the image of this webpage, can you determine the coordinates of the element I describe (with point)? This element acts as a navigation link, directing users to specific pages or resources upon activation, contributing to a seamless browsing experience within the Cloudflare webpage.",
"role": "user"
},
{
"content": "(28,20)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
seeclick-images/c08bda64dbbb1efe8bb7965ce7371730.png
|
<image>
Click on "Tips"
|
[
0.033520668745040894,
0.14795367419719696,
0.049498215317726135,
0.16279935836791992
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Click on "Tips"
|
[
79,
198,
117,
218
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nClick on \"Tips\"",
"role": "user"
},
{
"content": "(98,208)",
"role": "assistant"
}
] | 2,380
| 2,380
| 1,344
|
|
showui-web-images/showui_web_009223.jpg
|
<image>
Locate Careers
|
[
0.7274982333183289,
0.11720864474773407,
0.7596011757850647,
0.14907380938529968
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Locate Careers
|
[
1935,
173,
2020,
221
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nLocate Careers",
"role": "user"
},
{
"content": "(1977,197)",
"role": "assistant"
}
] | 2,683
| 2,660
| 1,484
|
|
doc_images/AmHHgw-Nep9dvzNYLRBvLVZqZLjV_3840x2160_CHAR_93_0.png
|
<image>
Given the following text: "/) directory. " , find the text in the document and click the 1th character "/" in the text.
|
[
0.23802083333333332,
0.45416666666666666,
0.240625,
0.4587962962962963
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Given the following text: "/) directory. " , find the text in the document and click the 1th character "/" in the text.
|
[
633,
673,
640,
680
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nGiven the following text: \"/) directory. \" , find the text in the document and click the 1th character \"/\" in the text.",
"role": "user"
},
{
"content": "(636,676)",
"role": "assistant"
}
] | 3,840
| 2,660
| 1,484
|
|
autogui-images/autogui_1301595_8d6ddf20.png
|
<image>
Based on my descriptions, find the locations of the mentioned element in this webpage screenshot (with point). This element provides access to a webpage featuring Apple accessories.
|
[
0.6569795608520508,
0.09578849375247955,
0.740260899066925,
0.16283586621284485
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Based on my descriptions, find the locations of the mentioned element in this webpage screenshot (with point). This element provides access to a webpage featuring Apple accessories.
|
[
367,
29,
414,
50
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nBased on my descriptions, find the locations of the mentioned element in this webpage screenshot (with point). This element provides access to a webpage featuring Apple accessories.",
"role": "user"
},
{
"content": "(390,39)",
"role": "assistant"
}
] | 560
| 560
| 308
|
|
autogui-images/autogui_350642_a320dff4.png
|
<image>
In this web page image, please locate the element as I describe it (with point). This element serves as a navigation tool, directing users to the Marketplace section of the website, where they can explore, purchase, and trade various digital items, such as the 'Samy The Cobra' featured on the page.
|
[
0.28324559330940247,
0.05433427542448044,
0.5249799489974976,
0.10375956445932388
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
In this web page image, please locate the element as I describe it (with point). This element serves as a navigation tool, directing users to the Marketplace section of the website, where they can explore, purchase, and trade various digital items, such as the 'Samy The Cobra' featured on the page.
|
[
63,
19,
117,
37
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nIn this web page image, please locate the element as I describe it (with point). This element serves as a navigation tool, directing users to the Marketplace section of the website, where they can explore, purchase, and trade various digital items, such as the 'Samy The Cobra' featured on the page.",
"role": "user"
},
{
"content": "(90,28)",
"role": "assistant"
}
] | 224
| 224
| 364
|
|
uground-images/uground_227343.jpg
|
<image>
Select Allow Cookies button
|
[
0.12493585050106049,
0.9709451794624329,
0.2759685814380646,
0.9929510354995728
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select Allow Cookies button
|
[
62,
978,
139,
1000
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect Allow Cookies button",
"role": "user"
},
{
"content": "(100,989)",
"role": "assistant"
}
] | 504
| 504
| 1,008
|
|
showui-web-images/showui_web_017434.jpg
|
<image>
Select View profile of George Fraser
|
[
0.574923038482666,
0.8815286755561829,
0.6333909034729004,
0.9046930074691772
] |
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
<image>
Select View profile of George Fraser
|
[
1497,
1283,
1649,
1317
] |
[
{
"content": "You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.\n\nOutput the coordinate pair exactly:\n(x,y)",
"role": "system"
},
{
"content": "<image>\nSelect View profile of George Fraser",
"role": "user"
},
{
"content": "(1573,1300)",
"role": "assistant"
}
] | 2,606
| 2,604
| 1,456
|
End of preview. Expand
in Data Studio
YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/datasets-cards)
easyr1-70k-hard-qwen7b-easy-gta1-4MP-no-resolution-in-prompt
This dataset was generated using the EasyR1 grounding dataset pipeline.
Generation Details
- Generated on: 2025-08-27 01:19:44 UTC
- Script:
push_easyr1_to_hf.py - Data directory:
/lustre/fsw/portfolios/nvr/users/aawadalla/LLaMA-Factory/data
Parameters Used
- Maximum samples: 70000
- Image resize (max megapixels): 4.0 MP
- Minimum native image resolution: 0.0 MP
- Prompt format:
gta1 - Output format:
coordinates - Random seed: 42
- Resampling enabled: False
- Icon upsampling ratio: Disabled (random sampling)
- pc-agent-e deduplication: False
- Debug images enabled: False
Dataset Groups
The following JSON/JSONL files were used to create this dataset:
Dataset Group 1
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/jedi-all-grounding-left_click-10px.jsonl
Dataset Group 2
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/os-atlas-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-3B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/os-atlas-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 3
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/pixmo-points-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/pixmo-points-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 4
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/autogui-grounding-only-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/autogui-grounding-only-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 5
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/seeclick-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/seeclick-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 6
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/pc-e-grounding-only-claude-instructions-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/pc-e-grounding-only-claude-instructions-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 7
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/omniact-grounding-only-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/omniact-grounding-only-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 8
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/showui-desktop-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/showui-desktop-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 9
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/showui-web-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/showui-web-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 10
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/uground-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/uground-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 11
Files (intersection of kept samples across all files):
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/waveui-qwen_tool_call-not_grounded-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/waveui-gta1-correctly_grounded-HelloKKMe_GTA1-7B-gta1.jsonl
Dataset Group 12
- /lustre/fs12/portfolios/nvr/projects/nvr_lacr_llm/users/aawadalla/grounding-data-filters/professional-apps-grounding-filtered-gta1-7b-converted-not_grounded-qwen2_5vl_7b-Qwen_Qwen2.5-VL-7B-Instruct-qwen_tool_call.jsonl
Dataset Statistics
- Total training samples: 68577
- Image dimensions: Variable
- Columns: image_path, prompt, normalized_bbox, images, easyr1_prompt, bbox, messages, original_image_width, image_width, image_height
System Prompt
The following system prompt is used for this dataset:
You are an expert UI element locator. Given a GUI image and a user's element description, provide the coordinates of the specified element as a single (x,y) point. For elements with area, return the center point.
Output the coordinate pair exactly:
(x,y)
Sample Entry
- User prompt:
Given the following text: "The quality of the data and annotation upper-bounds the quality of a downstream model. While there exist large text corpora and image-text pairs, high-quality video-text data is much harder to collect. First of all, manual labeling is more time-consuming, as it requires an annotator to watch an entire video. Second, videos have a temporal dimension, consisting of several scenes stacked together, and showing multiple actions. Accordingly, to establish a video dataset with high-quality captions, we propose an automatic approach leveraging multimodal inputs, such as textual video description, subtitles, and individual video frames. Specifically, we curate 3.8M high-resolution videos from the publicly available HD-VILA-100M dataset. We then split them into semantically consistent video clips, and apply multiple cross-modality teacher models to obtain captions for each video. Next, we finetune a retrieval model on a small subset where the best caption of each video is manually selected and then employ the model in the whole dataset to select the best caption as the annotation. In " , find the text in the document and click the 5th character "s" in the text.
- Assistant response: (852,730)
- Bounding box: [847, 725, 857, 736]
- Image path: doc_images/AmHHgw-Nep9dv3yPkTdG-MxD1Wka_1920x1080_CHAR_53_176.png
Usage
from datasets import load_dataset
dataset = load_dataset("mlfoundations-cua-dev/easyr1-70k-hard-qwen7b-easy-gta1-4MP-no-resolution-in-prompt")
# Access the training data
train_data = dataset['train']
# Example: Get the first sample
sample = train_data[0]
images = sample['images']
messages = sample['messages']
bbox = sample['bbox']
Prompt Formats
gta1
Standard GTA1 format without resolution information. Outputs coordinates in (x,y) format.
License
Please refer to the original dataset licenses for usage restrictions.
- Downloads last month
- 32