id
stringlengths 24
37
| image
imagewidth (px) 270
720
| conversations
listlengths 2
2
|
|---|---|---|
aitw_1684888717715898644_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Searchfor good French restaurants\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [LOCATION Icon]\nStep 2: Tap [X Icon]\nStep 3: Type [good French restaurants]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 5}}"
}
] |
|
aitw_1684888717715898644_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Searchfor good French restaurants\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [LOCATION Icon]\nStep 2: Tap [X Icon]\nStep 3: Type [good French restaurants]\nStep 4: Tap [Search here]\n\nCurrent screenshot is labeled with 28 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"Good french restaurants\"}"
}
] |
|
aitw_1684888717715898644_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Searchfor good French restaurants\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [LOCATION Icon]\nStep 2: Tap [X Icon]\nStep 3: Type [good French restaurants]\nStep 4: Tap [Search here]\nStep 5: Type [Good french restaurants]\n\nCurrent screenshot is labeled with 14 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 9}}"
}
] |
|
aitw_1684888717715898644_7
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Searchfor good French restaurants\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [LOCATION Icon]\nStep 2: Tap [X Icon]\nStep 3: Type [good French restaurants]\nStep 4: Tap [Search here]\nStep 5: Type [Good french restaurants]\nStep 6: Tap [good french restaurants near me]\n\nCurrent screenshot is labeled with 12 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 7) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_12917350239950899516_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Bangladesh?\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.55, 0.9]}}"
}
] |
|
aitw_12917350239950899516_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Bangladesh?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 12}}"
}
] |
|
aitw_12917350239950899516_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Bangladesh?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [Search apps, web, and more.]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"What's the news in Bangladesh?\"}"
}
] |
|
aitw_12917350239950899516_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Bangladesh?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [Search apps, web, and more.]\nStep 2: Type [What's the news in Bangladesh?]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.58, 0.64]}}"
}
] |
|
aitw_12917350239950899516_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Bangladesh?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [Search apps, web, and more.]\nStep 2: Type [What's the news in Bangladesh?]\nStep 3: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 22 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 14}}"
}
] |
|
aitw_12917350239950899516_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Bangladesh?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [Search apps, web, and more.]\nStep 2: Type [What's the news in Bangladesh?]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [what's the weather news in bangladesh]\n\nCurrent screenshot is labeled with 48 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_2139470499346320360_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.25, 0.91]}}"
}
] |
|
aitw_2139470499346320360_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.27, 0.91]}}"
}
] |
|
aitw_2139470499346320360_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.37, 0.89]}}"
}
] |
|
aitw_2139470499346320360_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 12}}"
}
] |
|
aitw_2139470499346320360_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [Search apps, web, and more..]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"flights from NYC to Buenos aires\"}"
}
] |
|
aitw_2139470499346320360_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [Search apps, web, and more..]\nStep 4: Type [flights from NYC to Buenos aires]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.5, 0.7]}}"
}
] |
|
aitw_2139470499346320360_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [Search apps, web, and more..]\nStep 4: Type [flights from NYC to Buenos aires]\nStep 5: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 16}}"
}
] |
|
aitw_2139470499346320360_7
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for flights from NYC to Buenos aires\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [Search apps, web, and more..]\nStep 4: Type [flights from NYC to Buenos aires]\nStep 5: Tap [somewhere on the screen]\nStep 6: Tap [flights from nyc to buenos aires]\n\nCurrent screenshot is labeled with 56 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 7) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_13988791135954828010_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Uruguay?\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"What's the news in Uruguay?\"}"
}
] |
|
aitw_13988791135954828010_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Uruguay?\n\nPrevious actions:\nStep 0: Type [What's the news in Uruguay?]\n\nCurrent screenshot is labeled with 8 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 7}}"
}
] |
|
aitw_13988791135954828010_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the news in Uruguay?\n\nPrevious actions:\nStep 0: Type [What's the news in Uruguay?]\nStep 1: Tap [today .]\n\nCurrent screenshot is labeled with 36 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_5621629839875730500_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the Instagram app\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SWIPE\": \"UP\"}"
}
] |
|
aitw_5621629839875730500_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the Instagram app\n\nPrevious actions:\nStep 0: Swipe [UP]\n\nCurrent screenshot is labeled with 42 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 7}}"
}
] |
|
aitw_5621629839875730500_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the Instagram app\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [SETTINGS Icon]\n\nCurrent screenshot is labeled with 19 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_1459760647753191721_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's a good restaurant in Houston?\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 9 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.28, 0.89]}}"
}
] |
|
aitw_1459760647753191721_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's a good restaurant in Houston?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 9 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.48, 0.89]}}"
}
] |
|
aitw_1459760647753191721_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's a good restaurant in Houston?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 24 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 13}}"
}
] |
|
aitw_1459760647753191721_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's a good restaurant in Houston?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more.]\n\nCurrent screenshot is labeled with 24 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"What's a good restaurant in Houston?\"}"
}
] |
|
aitw_1459760647753191721_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's a good restaurant in Houston?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more.]\nStep 3: Type [What's a good restaurant in Houston?]\n\nCurrent screenshot is labeled with 23 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 14}}"
}
] |
|
aitw_1459760647753191721_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's a good restaurant in Houston?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more.]\nStep 3: Type [What's a good restaurant in Houston?]\nStep 4: Tap [what's the best restaurant in houston]\n\nCurrent screenshot is labeled with 31 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_12237901697529872901_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for hotels in London\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 16 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"PRESS\": \"HOME\"}"
}
] |
|
aitw_12237901697529872901_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for hotels in London\n\nPrevious actions:\nStep 0: Press [HOME]\n\nCurrent screenshot is labeled with 11 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.52, 0.89]}}"
}
] |
|
aitw_12237901697529872901_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for hotels in London\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 24 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 13}}"
}
] |
|
aitw_12237901697529872901_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for hotels in London\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more.]\n\nCurrent screenshot is labeled with 24 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"hotels in London\"}"
}
] |
|
aitw_12237901697529872901_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for hotels in London\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more.]\nStep 3: Type [hotels in London]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 16}}"
}
] |
|
aitw_12237901697529872901_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for hotels in London\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more.]\nStep 3: Type [hotels in London]\nStep 4: Tap [hotels london in]\n\nCurrent screenshot is labeled with 33 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 20}}"
}
] |
|
aitw_12237901697529872901_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for hotels in London\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more.]\nStep 3: Type [hotels in London]\nStep 4: Tap [hotels london in]\nStep 5: Tap [Hotels London - in Best Price Guarantee]\n\nCurrent screenshot is labeled with 8 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_1597736946563356429_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 10 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SWIPE\": \"UP\"}"
}
] |
|
aitw_1597736946563356429_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\n\nCurrent screenshot is labeled with 39 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 3}}"
}
] |
|
aitw_1597736946563356429_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\n\nCurrent screenshot is labeled with 25 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 1}}"
}
] |
|
aitw_1597736946563356429_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\nStep 2: Tap [MAGNIFYING_GLASS Icon]\n\nCurrent screenshot is labeled with 32 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 18}}"
}
] |
|
aitw_1597736946563356429_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\nStep 2: Tap [MAGNIFYING_GLASS Icon]\nStep 3: Tap [Search YouTube]\n\nCurrent screenshot is labeled with 32 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 18}}"
}
] |
|
aitw_1597736946563356429_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\nStep 2: Tap [MAGNIFYING_GLASS Icon]\nStep 3: Tap [Search YouTube]\nStep 4: Tap [Search YouTube]\n\nCurrent screenshot is labeled with 32 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"latest video from GameSpot News\"}"
}
] |
|
aitw_1597736946563356429_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\nStep 2: Tap [MAGNIFYING_GLASS Icon]\nStep 3: Tap [Search YouTube]\nStep 4: Tap [Search YouTube]\nStep 5: Type [latest video from GameSpot News]\n\nCurrent screenshot is labeled with 8 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 7}}"
}
] |
|
aitw_1597736946563356429_7
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\nStep 2: Tap [MAGNIFYING_GLASS Icon]\nStep 3: Tap [Search YouTube]\nStep 4: Tap [Search YouTube]\nStep 5: Type [latest video from GameSpot News]\nStep 6: Tap [news ...]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 7) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SWIPE\": \"UP\"}"
}
] |
|
aitw_1597736946563356429_8
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\nStep 2: Tap [MAGNIFYING_GLASS Icon]\nStep 3: Tap [Search YouTube]\nStep 4: Tap [Search YouTube]\nStep 5: Type [latest video from GameSpot News]\nStep 6: Tap [news ...]\nStep 7: Swipe [UP]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 8) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.46, 0.41]}}"
}
] |
|
aitw_1597736946563356429_9
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the latest video from GameSpot News?\n\nPrevious actions:\nStep 0: Swipe [UP]\nStep 1: Tap [PLAY Icon]\nStep 2: Tap [MAGNIFYING_GLASS Icon]\nStep 3: Tap [Search YouTube]\nStep 4: Tap [Search YouTube]\nStep 5: Type [latest video from GameSpot News]\nStep 6: Tap [news ...]\nStep 7: Swipe [UP]\nStep 8: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 7 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 9) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_13631136942144223616_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: How much does a 2 bedroom apartment rent for in Portland?\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 10 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.52, 0.89]}}"
}
] |
|
aitw_13631136942144223616_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: How much does a 2 bedroom apartment rent for in Portland?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 10 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.42, 0.87]}}"
}
] |
|
aitw_13631136942144223616_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: How much does a 2 bedroom apartment rent for in Portland?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 10}}"
}
] |
|
aitw_13631136942144223616_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: How much does a 2 bedroom apartment rent for in Portland?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more..]\n\nCurrent screenshot is labeled with 22 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 11}}"
}
] |
|
aitw_13631136942144223616_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: How much does a 2 bedroom apartment rent for in Portland?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more..]\nStep 3: Tap [Search apps, web, and more.]\n\nCurrent screenshot is labeled with 22 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"How much does a 2 bedroom apartment rent for in Portland?\"}"
}
] |
|
aitw_13631136942144223616_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: How much does a 2 bedroom apartment rent for in Portland?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more..]\nStep 3: Tap [Search apps, web, and more.]\nStep 4: Type [How much does a 2 bedroom apartment rent for in Portland?]\n\nCurrent screenshot is labeled with 12 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 9}}"
}
] |
|
aitw_13631136942144223616_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: How much does a 2 bedroom apartment rent for in Portland?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [Search apps, web, and more..]\nStep 3: Tap [Search apps, web, and more.]\nStep 4: Type [How much does a 2 bedroom apartment rent for in Portland?]\nStep 5: Tap [...portland oregon]\n\nCurrent screenshot is labeled with 35 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_12735674421312260299_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 22 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"PRESS\": \"HOME\"}"
}
] |
|
aitw_12735674421312260299_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nStep 0: Press [HOME]\n\nCurrent screenshot is labeled with 10 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.34, 0.89]}}"
}
] |
|
aitw_12735674421312260299_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 12 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.32, 0.91]}}"
}
] |
|
aitw_12735674421312260299_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 12 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.33, 0.87]}}"
}
] |
|
aitw_12735674421312260299_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 10}}"
}
] |
|
aitw_12735674421312260299_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [Bearch apps, web, and more.]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"What's the weather like in London?\"}"
}
] |
|
aitw_12735674421312260299_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [Bearch apps, web, and more.]\nStep 5: Type [What's the weather like in London?]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 16}}"
}
] |
|
aitw_12735674421312260299_7
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's the weather like in London?\n\nPrevious actions:\nStep 0: Press [HOME]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [Bearch apps, web, and more.]\nStep 5: Type [What's the weather like in London?]\nStep 6: Tap [what's the weather like london in today]\n\nCurrent screenshot is labeled with 36 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 7) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_825030933088533866_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Set an alarm for 9am\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 9 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.86, 0.78]}}"
}
] |
|
aitw_825030933088533866_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Set an alarm for 9am\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 36 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 34}}"
}
] |
|
aitw_825030933088533866_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Set an alarm for 9am\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [8:00.mM]\n\nCurrent screenshot is labeled with 27 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 18}}"
}
] |
|
aitw_825030933088533866_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Set an alarm for 9am\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [8:00.mM]\nStep 2: Tap [9]\n\nCurrent screenshot is labeled with 28 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 13}}"
}
] |
|
aitw_825030933088533866_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Set an alarm for 9am\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [8:00.mM]\nStep 2: Tap [9]\nStep 3: Tap [9:00]\n\nCurrent screenshot is labeled with 28 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_15213977228389538034_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 14 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.43, 0.91]}}"
}
] |
|
aitw_15213977228389538034_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 21 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"BBQ restaurants\"}"
}
] |
|
aitw_15213977228389538034_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Type [BBQ restaurants]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"\"}"
}
] |
|
aitw_15213977228389538034_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Type [BBQ restaurants]\nStep 2: Type []\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 16}}"
}
] |
|
aitw_15213977228389538034_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Type [BBQ restaurants]\nStep 2: Type []\nStep 3: Tap [bbq restaurants near me]\n\nCurrent screenshot is labeled with 24 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 20}}"
}
] |
|
aitw_15213977228389538034_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Type [BBQ restaurants]\nStep 2: Type []\nStep 3: Tap [bbq restaurants near me]\nStep 4: Tap [Report this]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 16}}"
}
] |
|
aitw_15213977228389538034_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Type [BBQ restaurants]\nStep 2: Type []\nStep 3: Tap [bbq restaurants near me]\nStep 4: Tap [Report this]\nStep 5: Tap [bbq restaurants near me]\n\nCurrent screenshot is labeled with 24 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 21}}"
}
] |
|
aitw_15213977228389538034_7
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Type [BBQ restaurants]\nStep 2: Type []\nStep 3: Tap [bbq restaurants near me]\nStep 4: Tap [Report this]\nStep 5: Tap [bbq restaurants near me]\nStep 6: Tap [OK]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 7) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 16}}"
}
] |
|
aitw_15213977228389538034_8
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Search for good BBQ restaurants\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Type [BBQ restaurants]\nStep 2: Type []\nStep 3: Tap [bbq restaurants near me]\nStep 4: Tap [Report this]\nStep 5: Tap [bbq restaurants near me]\nStep 6: Tap [OK]\nStep 7: Tap [bbq restaurants near me]\n\nCurrent screenshot is labeled with 36 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 8) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_12826965904630740791_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 14 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.39, 0.88]}}"
}
] |
|
aitw_12826965904630740791_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.52, 0.92]}}"
}
] |
|
aitw_12826965904630740791_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.52, 0.88]}}"
}
] |
|
aitw_12826965904630740791_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.52, 0.88]}}"
}
] |
|
aitw_12826965904630740791_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.52, 0.54]}}"
}
] |
|
aitw_12826965904630740791_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 20 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.41, 0.9]}}"
}
] |
|
aitw_12826965904630740791_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [somewhere on the screen]\nStep 5: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 14 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.39, 0.9]}}"
}
] |
|
aitw_12826965904630740791_7
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [somewhere on the screen]\nStep 5: Tap [somewhere on the screen]\nStep 6: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 15 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 7) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.39, 0.9]}}"
}
] |
|
aitw_12826965904630740791_8
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [somewhere on the screen]\nStep 5: Tap [somewhere on the screen]\nStep 6: Tap [somewhere on the screen]\nStep 7: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 22 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 8) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TYPE\": \"menu at Taco Bell\"}"
}
] |
|
aitw_12826965904630740791_9
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [somewhere on the screen]\nStep 5: Tap [somewhere on the screen]\nStep 6: Tap [somewhere on the screen]\nStep 7: Tap [somewhere on the screen]\nStep 8: Type [menu at Taco Bell]\n\nCurrent screenshot is labeled with 26 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 9) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 16}}"
}
] |
|
aitw_12826965904630740791_10
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: What's on the menu at Taco Bell?\n\nPrevious actions:\nStep 0: Tap [somewhere on the screen]\nStep 1: Tap [somewhere on the screen]\nStep 2: Tap [somewhere on the screen]\nStep 3: Tap [somewhere on the screen]\nStep 4: Tap [somewhere on the screen]\nStep 5: Tap [somewhere on the screen]\nStep 6: Tap [somewhere on the screen]\nStep 7: Tap [somewhere on the screen]\nStep 8: Type [menu at Taco Bell]\nStep 9: Tap [menu at taco bell]\n\nCurrent screenshot is labeled with 31 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 10) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SET_TASK_STATUS\": \"COMPLETE\"}"
}
] |
|
aitw_15273831274038585554_0
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nNone\n\nCurrent screenshot is labeled with 14 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 0) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 0}}"
}
] |
|
aitw_15273831274038585554_1
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\n\nCurrent screenshot is labeled with 29 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 1) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 6}}"
}
] |
|
aitw_15273831274038585554_2
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\n\nCurrent screenshot is labeled with 31 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 2) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SWIPE\": \"DOWN\"}"
}
] |
|
aitw_15273831274038585554_3
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\n\nCurrent screenshot is labeled with 30 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 3) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 2}}"
}
] |
|
aitw_15273831274038585554_4
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\n\nCurrent screenshot is labeled with 30 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 4) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 2}}"
}
] |
|
aitw_15273831274038585554_5
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\n\nCurrent screenshot is labeled with 30 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 5) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 2}}"
}
] |
|
aitw_15273831274038585554_6
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\nStep 5: Tap [THREE_DOTS Icon]\n\nCurrent screenshot is labeled with 30 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 6) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 2}}"
}
] |
|
aitw_15273831274038585554_7
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\nStep 5: Tap [THREE_DOTS Icon]\nStep 6: Tap [THREE_DOTS Icon]\n\nCurrent screenshot is labeled with 31 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 7) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 3}}"
}
] |
|
aitw_15273831274038585554_8
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\nStep 5: Tap [THREE_DOTS Icon]\nStep 6: Tap [THREE_DOTS Icon]\nStep 7: Tap [THREE_DOTS Icon]\n\nCurrent screenshot is labeled with 29 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 8) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"NORMALIZED_COORDINATE\": [0.56, 0.19]}}"
}
] |
|
aitw_15273831274038585554_9
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\nStep 5: Tap [THREE_DOTS Icon]\nStep 6: Tap [THREE_DOTS Icon]\nStep 7: Tap [THREE_DOTS Icon]\nStep 8: Tap [somewhere on the screen]\n\nCurrent screenshot is labeled with 30 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 9) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"TAP\": {\"MARK\": 10}}"
}
] |
|
aitw_15273831274038585554_10
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\nStep 5: Tap [THREE_DOTS Icon]\nStep 6: Tap [THREE_DOTS Icon]\nStep 7: Tap [THREE_DOTS Icon]\nStep 8: Tap [somewhere on the screen]\nStep 9: Tap [PLAY Icon]\n\nCurrent screenshot is labeled with 39 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 10) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SWIPE\": \"UP\"}"
}
] |
|
aitw_15273831274038585554_11
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\nStep 5: Tap [THREE_DOTS Icon]\nStep 6: Tap [THREE_DOTS Icon]\nStep 7: Tap [THREE_DOTS Icon]\nStep 8: Tap [somewhere on the screen]\nStep 9: Tap [PLAY Icon]\nStep 10: Swipe [UP]\n\nCurrent screenshot is labeled with 36 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 11) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SWIPE\": \"UP\"}"
}
] |
|
aitw_15273831274038585554_12
|
[
{
"from": "user",
"value": "<image>\nImagine you are interacting with an Android device step by step. You can see the screen and know the previous actions through recorded history. The screenshots are labeled with bounding boxes and numeric marks to guide your interactions. Your task is to determine the next action to perform. The possible actions include tapping, swiping, navigating, typing, or setting the task status.\n\nUse the following format for your output, selecting the appropriate key-value pair:\n```json\n{\n \"TAP\": dict, // Use {\"MARK\": mark_index} where `mark_index` is an integer corresponding to the bounding box on the screen, or {\"NORMALIZED_COORDINATE\": [x, y]} where `x` and `y` are float numbers between 0 and 1 representing the normalized coordinates on the screen.\n \"SWIPE\": str, // Choose \"UP\", \"DOWN\", \"LEFT\", or \"RIGHT\"\n \"NAVIGATE\": str, // Choose \"BACK\", \"HOME\", or \"ENTER\"\n \"TYPE\": str, // The text to be typed\n \"SET_TASK_STATUS\": str, // Choose \"COMPLETE\" or \"IMPOSSIBLE\"\n}\n```\n\nCurrent task: Check the settings for the YouTube app\n\nPrevious actions:\nStep 0: Tap [PLAY Icon]\nStep 1: Tap [X Icon]\nStep 2: Swipe [DOWN]\nStep 3: Tap [THREE_DOTS Icon]\nStep 4: Tap [THREE_DOTS Icon]\nStep 5: Tap [THREE_DOTS Icon]\nStep 6: Tap [THREE_DOTS Icon]\nStep 7: Tap [THREE_DOTS Icon]\nStep 8: Tap [somewhere on the screen]\nStep 9: Tap [PLAY Icon]\nStep 10: Swipe [UP]\nStep 11: Swipe [UP]\n\nCurrent screenshot is labeled with 39 bounding boxes and numeric marks to guide your interactions. Determine your next action (Step 12) and provide the output in the specified dictionary format."
},
{
"from": "assistant",
"value": "{\"SWIPE\": \"DOWN\"}"
}
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.