messages
listlengths
2
200
images
listlengths
1
3
custom_id
int64
0
66.2k
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Type \"panasonic triple\" in the search box to start the search for the desired item on eBay.\nAction: {\"action\": \"TYPE\", \"value\": \"panasonic triple \", \"position\": null}<image>Thought: Press enter to execute the search for \"panasonic triple a\" on eBay, generating relevant results.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on the first entry to select the Panasonic product.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [78, 584]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_7.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_8.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_9.png" ]
7,900
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Press enter to execute the search for \"panasonic triple a\" on eBay, generating relevant results.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}<image>Thought: Click on the first entry to select the Panasonic product.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [78, 584]}<image>", "role": "user" }, { "content": "Thought: Scroll down to view more product details or other listings related to Panasonic Triple A.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_8.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_9.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_10.png" ]
7,901
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Click on the first entry to select the Panasonic product.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [78, 584]}<image>Thought: Scroll down to view more product details or other listings related to Panasonic Triple A.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click \"Add to cart\" to include the Panasonic item in your shopping cart for later purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [253, 721]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_9.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_10.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_11.png" ]
7,902
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Scroll down to view more product details or other listings related to Panasonic Triple A.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>Thought: Click \"Add to cart\" to include the Panasonic item in your shopping cart for later purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [253, 721]}<image>", "role": "user" }, { "content": "Thought: Click the \"View in cart\" button to proceed to the shopping cart and review added items.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [254, 739]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_10.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_11.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_12.png" ]
7,903
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Click \"Add to cart\" to include the Panasonic item in your shopping cart for later purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [253, 721]}<image>Thought: Click the \"View in cart\" button to proceed to the shopping cart and review added items.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [254, 739]}<image>", "role": "user" }, { "content": "Thought: Click the \"View in cart\" option to proceed with viewing the items selected for purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [294, 907]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_11.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_12.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_13.png" ]
7,904
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Click the \"View in cart\" button to proceed to the shopping cart and review added items.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [254, 739]}<image>Thought: Click the \"View in cart\" option to proceed with viewing the items selected for purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [294, 907]}<image>", "role": "user" }, { "content": "Thought: Click \"View cart\" to proceed with reviewing the items added before checkout.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [262, 741]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_12.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_13.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_14.png" ]
7,905
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Click the \"View in cart\" option to proceed with viewing the items selected for purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [294, 907]}<image>Thought: Click \"View cart\" to proceed with reviewing the items added before checkout.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [262, 741]}<image>", "role": "user" }, { "content": "Thought: Click the \"Go to checkout\" button to proceed with the purchase of items in the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [224, 436]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_13.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_14.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_15.png" ]
7,906
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for panasonic triple a on ebay.com, select the first entry, add it to the cart, then select checkout.<image>Thought: Click \"View cart\" to proceed with reviewing the items added before checkout.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [262, 741]}<image>Thought: Click the \"Go to checkout\" button to proceed with the purchase of items in the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [224, 436]}<image>", "role": "user" }, { "content": "Thought: status task complete indicates the user successfully completed their purchase on eBay after signing in.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/15564804487625883335_14.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_15.png", "training_data/AITW/aitw_images/webshopping/15564804487625883335_16.png" ]
7,907
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add usb-c to usb-b to the cart on newegg, then select checkout.<image>", "role": "user" }, { "content": "Thought: Open the Newegg app to begin shopping for the USB-C to USB-B cable. Clicking on the CHAT icon indicates you may want assistance or have questions before proceeding.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [124, 532]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/918690821205618722_0.png" ]
7,908
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add usb-c to usb-b to the cart on newegg, then select checkout.<image>Thought: Open the Newegg app to begin shopping for the USB-C to USB-B cable. Clicking on the CHAT icon indicates you may want assistance or have questions before proceeding.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [124, 532]}<image>", "role": "user" }, { "content": "Thought: Type \"newegg\" in the search bar to find the website for shopping.\nAction: {\"action\": \"TYPE\", \"value\": \"newegg\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/918690821205618722_0.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_1.png" ]
7,909
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add usb-c to usb-b to the cart on newegg, then select checkout.<image>Thought: Open the Newegg app to begin shopping for the USB-C to USB-B cable. Clicking on the CHAT icon indicates you may want assistance or have questions before proceeding.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [124, 532]}<image>Thought: Type \"newegg\" in the search bar to find the website for shopping.\nAction: {\"action\": \"TYPE\", \"value\": \"newegg\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on \"Newegg\" in the search results to navigate to the website and begin shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [90, 101]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/918690821205618722_0.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_1.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_2.png" ]
7,910
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add usb-c to usb-b to the cart on newegg, then select checkout.<image>Thought: Type \"newegg\" in the search bar to find the website for shopping.\nAction: {\"action\": \"TYPE\", \"value\": \"newegg\", \"position\": null}<image>Thought: Click on \"Newegg\" in the search results to navigate to the website and begin shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [90, 101]}<image>", "role": "user" }, { "content": "Thought: Click on Newegg to access the official site for product search and purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [32, 190]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/918690821205618722_1.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_2.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_3.png" ]
7,911
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add usb-c to usb-b to the cart on newegg, then select checkout.<image>Thought: Click on \"Newegg\" in the search results to navigate to the website and begin shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [90, 101]}<image>Thought: Click on Newegg to access the official site for product search and purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [32, 190]}<image>", "role": "user" }, { "content": "Thought: Click the magnifying glass to open the search feature for finding USB-C to USB-B products.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [170, 76]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/918690821205618722_2.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_3.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_4.png" ]
7,912
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add usb-c to usb-b to the cart on newegg, then select checkout.<image>Thought: Click on Newegg to access the official site for product search and purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [32, 190]}<image>Thought: Click the magnifying glass to open the search feature for finding USB-C to USB-B products.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [170, 76]}<image>", "role": "user" }, { "content": "Thought: Select the correct country from the \"COUNTRIES & REGIONS\" dropdown to continue shopping on Newegg.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/918690821205618722_3.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_4.png", "training_data/AITW/aitw_images/webshopping/918690821205618722_5.png" ]
7,913
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>", "role": "user" }, { "content": "Thought: Click the magnifying glass icon to initiate a search for 4K TVs on Best Buy.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [366, 86]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_0.png" ]
7,914
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click the magnifying glass icon to initiate a search for 4K TVs on Best Buy.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [366, 86]}<image>", "role": "user" }, { "content": "Thought: Press home to exit the current app and return to the home screen for efficient navigation.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_0.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_1.png" ]
7,915
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click the magnifying glass icon to initiate a search for 4K TVs on Best Buy.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [366, 86]}<image>Thought: Press home to exit the current app and return to the home screen for efficient navigation.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Open Chrome to begin searching for 4K TVs on Best Buy.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [332, 178]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_0.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_1.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_2.png" ]
7,916
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Press home to exit the current app and return to the home screen for efficient navigation.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>Thought: Open Chrome to begin searching for 4K TVs on Best Buy.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [332, 178]}<image>", "role": "user" }, { "content": "Thought: You pressed home likely to return to the main screen and access the Best Buy app or website for your search.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_1.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_2.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_3.png" ]
7,917
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Open Chrome to begin searching for 4K TVs on Best Buy.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [332, 178]}<image>Thought: You pressed home likely to return to the main screen and access the Best Buy app or website for your search.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Scroll down to reveal more apps on the home screen and locate the Best Buy app.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_2.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_3.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_4.png" ]
7,918
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: You pressed home likely to return to the main screen and access the Best Buy app or website for your search.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>Thought: Scroll down to reveal more apps on the home screen and locate the Best Buy app.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click Chrome to open the web browser for searching the best value 4K TV.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [216, 440]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_3.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_4.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_5.png" ]
7,919
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Scroll down to reveal more apps on the home screen and locate the Best Buy app.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>Thought: Click Chrome to open the web browser for searching the best value 4K TV.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [216, 440]}<image>", "role": "user" }, { "content": "Thought: Press back to return from the current screen to refine the search for a 4K TV.\nAction: {\"action\": \"PRESS BACK\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_4.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_5.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_6.png" ]
7,920
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click Chrome to open the web browser for searching the best value 4K TV.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [216, 440]}<image>Thought: Press back to return from the current screen to refine the search for a 4K TV.\nAction: {\"action\": \"PRESS BACK\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: You want to start your search for a 4K TV, so you click on \"rget.com\" to access the website.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [194, 96]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_5.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_6.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_7.png" ]
7,921
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Press back to return from the current screen to refine the search for a 4K TV.\nAction: {\"action\": \"PRESS BACK\", \"value\": null, \"position\": null}<image>Thought: You want to start your search for a 4K TV, so you click on \"rget.com\" to access the website.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [194, 96]}<image>", "role": "user" }, { "content": "Thought: Think to initiate a search by typing \"Best Buy\" into the search bar at the top.\nAction: {\"action\": \"TYPE\", \"value\": \"Best Buy\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_6.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_7.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_8.png" ]
7,922
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: You want to start your search for a 4K TV, so you click on \"rget.com\" to access the website.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [194, 96]}<image>Thought: Think to initiate a search by typing \"Best Buy\" into the search bar at the top.\nAction: {\"action\": \"TYPE\", \"value\": \"Best Buy\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on \"Best\" to select the Best Buy link for accessing their website.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [133, 176]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_7.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_8.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_9.png" ]
7,923
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Think to initiate a search by typing \"Best Buy\" into the search bar at the top.\nAction: {\"action\": \"TYPE\", \"value\": \"Best Buy\", \"position\": null}<image>Thought: Click on \"Best\" to select the Best Buy link for accessing their website.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [133, 176]}<image>", "role": "user" }, { "content": "Thought: Scroll down to view more options and results related to 4K TVs on Best Buy.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_8.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_9.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_10.png" ]
7,924
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click on \"Best\" to select the Best Buy link for accessing their website.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [133, 176]}<image>Thought: Scroll down to view more options and results related to 4K TVs on Best Buy.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on the \"Buy®\" button to navigate to the Best Buy website for shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [114, 224]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_9.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_10.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_11.png" ]
7,925
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Scroll down to view more options and results related to 4K TVs on Best Buy.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>Thought: Click on the \"Buy®\" button to navigate to the Best Buy website for shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [114, 224]}<image>", "role": "user" }, { "content": "Thought: Click on \"Open in new tab\" to view the Best Buy page without navigating away from the current page.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [256, 280]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_10.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_11.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_12.png" ]
7,926
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click on the \"Buy®\" button to navigate to the Best Buy website for shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [114, 224]}<image>Thought: Click on \"Open in new tab\" to view the Best Buy page without navigating away from the current page.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [256, 280]}<image>", "role": "user" }, { "content": "Thought: Click on the \"Best Buy\" link to access their official site for searching 4K TVs.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [71, 213]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_11.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_12.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_13.png" ]
7,927
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click on \"Open in new tab\" to view the Best Buy page without navigating away from the current page.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [256, 280]}<image>Thought: Click on the \"Best Buy\" link to access their official site for searching 4K TVs.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [71, 213]}<image>", "role": "user" }, { "content": "Thought: Click on tbuy.com to access the Best Buy website for shopping options.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [165, 87]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_12.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_13.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_14.png" ]
7,928
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click on the \"Best Buy\" link to access their official site for searching 4K TVs.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [71, 213]}<image>Thought: Click on tbuy.com to access the Best Buy website for shopping options.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [165, 87]}<image>", "role": "user" }, { "content": "Thought: Press back to return to the previous page and refine the search for a better 4K TV.\nAction: {\"action\": \"PRESS BACK\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_13.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_14.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_15.png" ]
7,929
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Click on tbuy.com to access the Best Buy website for shopping options.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [165, 87]}<image>Thought: Press back to return to the previous page and refine the search for a better 4K TV.\nAction: {\"action\": \"PRESS BACK\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click Español to select the Spanish language for better understanding while searching for 4K TVs.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [390, 149]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_14.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_15.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_16.png" ]
7,930
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for the best value 4K TV on Best Buy<image>Thought: Press back to return to the previous page and refine the search for a better 4K TV.\nAction: {\"action\": \"PRESS BACK\", \"value\": null, \"position\": null}<image>Thought: Click Español to select the Spanish language for better understanding while searching for 4K TVs.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [390, 149]}<image>", "role": "user" }, { "content": "Thought: You have successfully navigated to the Best Buy website to search for TVs.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10588022362023697645_15.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_16.png", "training_data/AITW/aitw_images/webshopping/10588022362023697645_17.png" ]
7,931
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>", "role": "user" }, { "content": "Thought: Press home to exit the current application and return to the main screen.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_0.png" ]
7,932
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Press home to exit the current application and return to the main screen.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click the ENVELOPE icon to access email notifications regarding your order on target.com or related issues.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [496, 1139]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_0.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_1.png" ]
7,933
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Press home to exit the current application and return to the main screen.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>Thought: Click the ENVELOPE icon to access email notifications regarding your order on target.com or related issues.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [496, 1139]}<image>", "role": "user" }, { "content": "Thought: Click on the item link to access the product page for \"macbook pro 15 inch.\"\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [420, 62]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_0.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_1.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_2.png" ]
7,934
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Click the ENVELOPE icon to access email notifications regarding your order on target.com or related issues.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [496, 1139]}<image>Thought: Click on the item link to access the product page for \"macbook pro 15 inch.\"\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [420, 62]}<image>", "role": "user" }, { "content": "Thought: Type \"target.com\" in the search bar to navigate to Target’s website.\nAction: {\"action\": \"TYPE\", \"value\": \"target.com\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_1.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_2.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_3.png" ]
7,935
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Click on the item link to access the product page for \"macbook pro 15 inch.\"\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [420, 62]}<image>Thought: Type \"target.com\" in the search bar to navigate to Target’s website.\nAction: {\"action\": \"TYPE\", \"value\": \"target.com\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on \"Target: Expect More\" to navigate to the Target website where the product can be added to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [135, 176]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_2.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_3.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_4.png" ]
7,936
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Type \"target.com\" in the search bar to navigate to Target’s website.\nAction: {\"action\": \"TYPE\", \"value\": \"target.com\", \"position\": null}<image>Thought: Click on \"Target: Expect More\" to navigate to the Target website where the product can be added to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [135, 176]}<image>", "role": "user" }, { "content": "Thought: Click the Search bar to access the keyboard for inputting \"macbook pro 15 inch.\"\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [184, 366]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_3.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_4.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_5.png" ]
7,937
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Click on \"Target: Expect More\" to navigate to the Target website where the product can be added to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [135, 176]}<image>Thought: Click the Search bar to access the keyboard for inputting \"macbook pro 15 inch.\"\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [184, 366]}<image>", "role": "user" }, { "content": "Thought: Type \"macbook pro 15 inch\" in the search bar to find the desired product on target.com.\nAction: {\"action\": \"TYPE\", \"value\": \"macbook pro 15 inch\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_4.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_5.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_6.png" ]
7,938
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Click the Search bar to access the keyboard for inputting \"macbook pro 15 inch.\"\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [184, 366]}<image>Thought: Type \"macbook pro 15 inch\" in the search bar to find the desired product on target.com.\nAction: {\"action\": \"TYPE\", \"value\": \"macbook pro 15 inch\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click the magnifying glass to search for \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [580, 192]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_5.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_6.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_7.png" ]
7,939
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Type \"macbook pro 15 inch\" in the search bar to find the desired product on target.com.\nAction: {\"action\": \"TYPE\", \"value\": \"macbook pro 15 inch\", \"position\": null}<image>Thought: Click the magnifying glass to search for \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [580, 192]}<image>", "role": "user" }, { "content": "Thought: Scroll down to find the product listing for \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_6.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_7.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_8.png" ]
7,940
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Click the magnifying glass to search for \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [580, 192]}<image>Thought: Scroll down to find the product listing for \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: You click on \"Targus Magnetic Privacy\" to access the product details for adding it to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [400, 606]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_7.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_8.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_9.png" ]
7,941
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Scroll down to find the product listing for \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>Thought: You click on \"Targus Magnetic Privacy\" to access the product details for adding it to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [400, 606]}<image>", "role": "user" }, { "content": "Thought: Scroll down to view more options or products related to \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_8.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_9.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_10.png" ]
7,942
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: You click on \"Targus Magnetic Privacy\" to access the product details for adding it to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [400, 606]}<image>Thought: Scroll down to view more options or products related to \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click \"Add to cart\" to confirm my desire to purchase the MacBook Pro 15 inch.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [406, 1047]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_9.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_10.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_11.png" ]
7,943
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Scroll down to view more options or products related to \"macbook pro 15 inch\" on target.com.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>Thought: Click \"Add to cart\" to confirm my desire to purchase the MacBook Pro 15 inch.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [406, 1047]}<image>", "role": "user" }, { "content": "Thought: Click the \"View cart & check out\" button to proceed with the checkout process for the added item.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [361, 685]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_10.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_11.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_12.png" ]
7,944
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Click \"Add to cart\" to confirm my desire to purchase the MacBook Pro 15 inch.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [406, 1047]}<image>Thought: Click the \"View cart & check out\" button to proceed with the checkout process for the added item.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [361, 685]}<image>", "role": "user" }, { "content": "Thought: Click on \"checkout\" to proceed with purchasing the items in the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [344, 1302]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_11.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_12.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_13.png" ]
7,945
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add \"macbook pro 15 inch\" to the cart on target.com, then select checkout.<image>Thought: Click the \"View cart & check out\" button to proceed with the checkout process for the added item.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [361, 685]}<image>Thought: Click on \"checkout\" to proceed with purchasing the items in the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [344, 1302]}<image>", "role": "user" }, { "content": "Thought: Complete the checkout process after adding the MacBook Pro to confirm purchase.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10273048359718610594_12.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_13.png", "training_data/AITW/aitw_images/webshopping/10273048359718610594_14.png" ]
7,946
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>", "role": "user" }, { "content": "Thought: Click the back button to return to the previous screen, likely for a better view of options.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [252, 1373]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_0.png" ]
7,947
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>Thought: Click the back button to return to the previous screen, likely for a better view of options.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [252, 1373]}<image>", "role": "user" }, { "content": "Thought: Click Walmart to access the website where I can view and add products to my shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [449, 471]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_0.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_1.png" ]
7,948
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>Thought: Click the back button to return to the previous screen, likely for a better view of options.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [252, 1373]}<image>Thought: Click Walmart to access the website where I can view and add products to my shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [449, 471]}<image>", "role": "user" }, { "content": "Thought: Click the walmart.com link to access the Walmart website for shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [199, 682]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_0.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_1.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_2.png" ]
7,949
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>Thought: Click Walmart to access the website where I can view and add products to my shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [449, 471]}<image>Thought: Click the walmart.com link to access the Walmart website for shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [199, 682]}<image>", "role": "user" }, { "content": "Thought: Click on the search bar to input the product name \"razer nari\" for the search function.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [431, 210]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_1.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_2.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_3.png" ]
7,950
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>Thought: Click the walmart.com link to access the Walmart website for shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [199, 682]}<image>Thought: Click on the search bar to input the product name \"razer nari\" for the search function.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [431, 210]}<image>", "role": "user" }, { "content": "Thought: You type \"razer nari\" to search for the product in the search bar of Walmart's website.\nAction: {\"action\": \"TYPE\", \"value\": \"razer nari\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_2.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_3.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_4.png" ]
7,951
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>Thought: Click on the search bar to input the product name \"razer nari\" for the search function.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [431, 210]}<image>Thought: You type \"razer nari\" to search for the product in the search bar of Walmart's website.\nAction: {\"action\": \"TYPE\", \"value\": \"razer nari\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click the search bar to enter the query \"razer nari\" and view relevant product results.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [545, 222]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_3.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_4.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_5.png" ]
7,952
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>Thought: You type \"razer nari\" to search for the product in the search bar of Walmart's website.\nAction: {\"action\": \"TYPE\", \"value\": \"razer nari\", \"position\": null}<image>Thought: Click the search bar to enter the query \"razer nari\" and view relevant product results.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [545, 222]}<image>", "role": "user" }, { "content": "Thought: Click \"Add\" on the first entry to add the selected razer nari headset to the shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [614, 1075]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_4.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_5.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_6.png" ]
7,953
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: View the shopping cart on walmart.com. Search for \"razer nari\" on walmart.com, select the first entry, and add it to the cart.<image>Thought: Click the search bar to enter the query \"razer nari\" and view relevant product results.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [545, 222]}<image>Thought: Click \"Add\" on the first entry to add the selected razer nari headset to the shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [614, 1075]}<image>", "role": "user" }, { "content": "Thought: You successfully viewed your shopping cart by completing the search and adding an item, confirming task completion.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/1236520592426024415_5.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_6.png", "training_data/AITW/aitw_images/webshopping/1236520592426024415_7.png" ]
7,954
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>", "role": "user" }, { "content": "Thought: Press home to exit the current settings and return to the main screen for easier access to Newegg.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_0.png" ]
7,955
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Press home to exit the current settings and return to the main screen for easier access to Newegg.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click the ENVELOPE to open the Gmail app for accessing order details or confirmations related to Newegg.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [253, 532]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_0.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_1.png" ]
7,956
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Press home to exit the current settings and return to the main screen for easier access to Newegg.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>Thought: Click the ENVELOPE to open the Gmail app for accessing order details or confirmations related to Newegg.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [253, 532]}<image>", "role": "user" }, { "content": "Thought: You need to search for \"Logitech G502\" on a shopping platform. Click the search bar labeled \"macbook\" to enter your search query.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [211, 43]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_0.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_1.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_2.png" ]
7,957
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click the ENVELOPE to open the Gmail app for accessing order details or confirmations related to Newegg.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [253, 532]}<image>Thought: You need to search for \"Logitech G502\" on a shopping platform. Click the search bar labeled \"macbook\" to enter your search query.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [211, 43]}<image>", "role": "user" }, { "content": "Thought: Type \"newegg\" in the search bar to find the website.\nAction: {\"action\": \"TYPE\", \"value\": \"newegg\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_1.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_2.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_3.png" ]
7,958
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: You need to search for \"Logitech G502\" on a shopping platform. Click the search bar labeled \"macbook\" to enter your search query.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [211, 43]}<image>Thought: Type \"newegg\" in the search bar to find the website.\nAction: {\"action\": \"TYPE\", \"value\": \"newegg\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click Newegg to access the online store for e-commerce shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [96, 212]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_2.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_3.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_4.png" ]
7,959
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Type \"newegg\" in the search bar to find the website.\nAction: {\"action\": \"TYPE\", \"value\": \"newegg\", \"position\": null}<image>Thought: Click Newegg to access the online store for e-commerce shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [96, 212]}<image>", "role": "user" }, { "content": "Thought: Click on \"Computer parts\" to navigate to the relevant category for adding the Logitech G502.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [146, 564]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_3.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_4.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_5.png" ]
7,960
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click Newegg to access the online store for e-commerce shopping.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [96, 212]}<image>Thought: Click on \"Computer parts\" to navigate to the relevant category for adding the Logitech G502.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [146, 564]}<image>", "role": "user" }, { "content": "Thought: Click the magnifying glass to access the search feature for finding the Logitech G502.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [173, 135]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_4.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_5.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_6.png" ]
7,961
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click on \"Computer parts\" to navigate to the relevant category for adding the Logitech G502.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [146, 564]}<image>Thought: Click the magnifying glass to access the search feature for finding the Logitech G502.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [173, 135]}<image>", "role": "user" }, { "content": "Thought: Click on HOT SEARCH to access trending products related to your interest in the Logitech G502 mouse.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [115, 128]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_5.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_6.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_7.png" ]
7,962
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click the magnifying glass to access the search feature for finding the Logitech G502.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [173, 135]}<image>Thought: Click on HOT SEARCH to access trending products related to your interest in the Logitech G502 mouse.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [115, 128]}<image>", "role": "user" }, { "content": "Thought: Type \"logitech g502\" in the search bar to find the product on Newegg's website.\nAction: {\"action\": \"TYPE\", \"value\": \"logitech g502\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_6.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_7.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_8.png" ]
7,963
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click on HOT SEARCH to access trending products related to your interest in the Logitech G502 mouse.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [115, 128]}<image>Thought: Type \"logitech g502\" in the search bar to find the product on Newegg's website.\nAction: {\"action\": \"TYPE\", \"value\": \"logitech g502\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on the logitech g502 to view its details and add it to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [76, 380]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_7.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_8.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_9.png" ]
7,964
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Type \"logitech g502\" in the search bar to find the product on Newegg's website.\nAction: {\"action\": \"TYPE\", \"value\": \"logitech g502\", \"position\": null}<image>Thought: Click on the logitech g502 to view its details and add it to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [76, 380]}<image>", "role": "user" }, { "content": "Thought: Scroll down to view more search results for the Logitech G502 on Newegg.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_8.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_9.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_10.png" ]
7,965
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click on the logitech g502 to view its details and add it to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [76, 380]}<image>Thought: Scroll down to view more search results for the Logitech G502 on Newegg.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Scroll up to view the results of the search for \"logitech g502.\"\nAction: {\"action\": \"SCROLL UP\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_9.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_10.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_11.png" ]
7,966
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Scroll down to view more search results for the Logitech G502 on Newegg.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>Thought: Scroll up to view the results of the search for \"logitech g502.\"\nAction: {\"action\": \"SCROLL UP\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on the Logitech G502 HERO to view its details and add it to my cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [319, 458]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_10.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_11.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_12.png" ]
7,967
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Scroll up to view the results of the search for \"logitech g502.\"\nAction: {\"action\": \"SCROLL UP\", \"value\": null, \"position\": null}<image>Thought: Click on the Logitech G502 HERO to view its details and add it to my cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [319, 458]}<image>", "role": "user" }, { "content": "Thought: Scroll down to view more details about the Logitech G502 mouse and locate the \"ADD TO CART\" button.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_11.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_12.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_13.png" ]
7,968
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click on the Logitech G502 HERO to view its details and add it to my cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [319, 458]}<image>Thought: Scroll down to view more details about the Logitech G502 mouse and locate the \"ADD TO CART\" button.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click the \"ADD TO CART\" button to quickly add Logitech G502 to my shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [209, 640]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_12.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_13.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_14.png" ]
7,969
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Scroll down to view more details about the Logitech G502 mouse and locate the \"ADD TO CART\" button.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>Thought: Click the \"ADD TO CART\" button to quickly add Logitech G502 to my shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [209, 640]}<image>", "role": "user" }, { "content": "Thought: Click \"NO, THANKS\" to skip the protection plan and proceed with adding the Logitech G502 to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [121, 634]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_13.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_14.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_15.png" ]
7,970
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click the \"ADD TO CART\" button to quickly add Logitech G502 to my shopping cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [209, 640]}<image>Thought: Click \"NO, THANKS\" to skip the protection plan and proceed with adding the Logitech G502 to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [121, 634]}<image>", "role": "user" }, { "content": "Thought: Click on the SHOPPING CART to view items added for purchase confirmation.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [30, 641]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_14.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_15.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_16.png" ]
7,971
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Add logitech g502 to the cart on newegg<image>Thought: Click \"NO, THANKS\" to skip the protection plan and proceed with adding the Logitech G502 to the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [121, 634]}<image>Thought: Click on the SHOPPING CART to view items added for purchase confirmation.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [30, 641]}<image>", "role": "user" }, { "content": "Thought: You would have navigated to the cart to confirm that the Logitech G502 was successfully added. Seeing \"Cart (2 Items)\" indicates that the item count has increased, confirming completion.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10216814946811914736_15.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_16.png", "training_data/AITW/aitw_images/webshopping/10216814946811914736_17.png" ]
7,972
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for 35 best books of all time<image>", "role": "user" }, { "content": "Thought: Press home to exit the email app and return to the main screen for searching books.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/12648952919912619913_0.png" ]
7,973
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for 35 best books of all time<image>Thought: Press home to exit the email app and return to the main screen for searching books.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click the Google Chrome icon to open the browser for searching for books.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [362, 1032]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/12648952919912619913_0.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_1.png" ]
7,974
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for 35 best books of all time<image>Thought: Press home to exit the email app and return to the main screen for searching books.\nAction: {\"action\": \"PRESS HOME\", \"value\": null, \"position\": null}<image>Thought: Click the Google Chrome icon to open the browser for searching for books.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [362, 1032]}<image>", "role": "user" }, { "content": "Thought: Press enter to submit the search query in the search bar for best books.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/12648952919912619913_0.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_1.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_2.png" ]
7,975
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for 35 best books of all time<image>Thought: Click the Google Chrome icon to open the browser for searching for books.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [362, 1032]}<image>Thought: Press enter to submit the search query in the search bar for best books.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Type in the search bar to find information on the best books of all time.\nAction: {\"action\": \"TYPE\", \"value\": \"35 best books of all time\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/12648952919912619913_1.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_2.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_3.png" ]
7,976
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for 35 best books of all time<image>Thought: Press enter to submit the search query in the search bar for best books.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}<image>Thought: Type in the search bar to find information on the best books of all time.\nAction: {\"action\": \"TYPE\", \"value\": \"35 best books of all time\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Press enter to execute the search for \"35 best books of all time\" in the search bar.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/12648952919912619913_2.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_3.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_4.png" ]
7,977
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for 35 best books of all time<image>Thought: Type in the search bar to find information on the best books of all time.\nAction: {\"action\": \"TYPE\", \"value\": \"35 best books of all time\", \"position\": null}<image>Thought: Press enter to execute the search for \"35 best books of all time\" in the search bar.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on the highlighted link to access Amazon's reading section for book recommendations.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [254, 384]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/12648952919912619913_3.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_4.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_5.png" ]
7,978
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Search for 35 best books of all time<image>Thought: Press enter to execute the search for \"35 best books of all time\" in the search bar.\nAction: {\"action\": \"PRESS ENTER\", \"value\": null, \"position\": null}<image>Thought: Click on the highlighted link to access Amazon's reading section for book recommendations.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [254, 384]}<image>", "role": "user" }, { "content": "Thought: You would have typed in your search query for the best books in the search bar. After finding relevant results, you successfully completed the task.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/12648952919912619913_4.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_5.png", "training_data/AITW/aitw_images/webshopping/12648952919912619913_6.png" ]
7,979
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>", "role": "user" }, { "content": "Thought: Click GOOGLE to open the browser and search for 4K TVs on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [35, 364]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_0.png" ]
7,980
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>Thought: Click GOOGLE to open the browser and search for 4K TVs on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [35, 364]}<image>", "role": "user" }, { "content": "Thought: Status task impossible; the search term does not relate to finding a best-rated 4K TV on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [205, 78]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_0.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_1.png" ]
7,981
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>Thought: Click GOOGLE to open the browser and search for 4K TVs on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [35, 364]}<image>Thought: Status task impossible; the search term does not relate to finding a best-rated 4K TV on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [205, 78]}<image>", "role": "user" }, { "content": "Thought: You click X to remove the current search query related to wallets.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [249, 34]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_0.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_1.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_2.png" ]
7,982
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>Thought: Status task impossible; the search term does not relate to finding a best-rated 4K TV on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [205, 78]}<image>Thought: You click X to remove the current search query related to wallets.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [249, 34]}<image>", "role": "user" }, { "content": "Thought: You click \"Search\" to initiate the query for the best rated 4K TV on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [114, 39]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_1.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_2.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_3.png" ]
7,983
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>Thought: You click X to remove the current search query related to wallets.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [249, 34]}<image>Thought: You click \"Search\" to initiate the query for the best rated 4K TV on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [114, 39]}<image>", "role": "user" }, { "content": "Thought: Type \"the best rated 4K TV on Walmart\" to initiate a search for the desired product in the search bar.\nAction: {\"action\": \"TYPE\", \"value\": \"the best rated 4K TV on Walmart \", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_2.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_3.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_4.png" ]
7,984
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>Thought: You click \"Search\" to initiate the query for the best rated 4K TV on Walmart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [114, 39]}<image>Thought: Type \"the best rated 4K TV on Walmart\" to initiate a search for the desired product in the search bar.\nAction: {\"action\": \"TYPE\", \"value\": \"the best rated 4K TV on Walmart \", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on \"walmart 2022\" to access relevant content about Walmart's products from that year.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [144, 86]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_3.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_4.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_5.png" ]
7,985
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>Thought: Type \"the best rated 4K TV on Walmart\" to initiate a search for the desired product in the search bar.\nAction: {\"action\": \"TYPE\", \"value\": \"the best rated 4K TV on Walmart \", \"position\": null}<image>Thought: Click on \"walmart 2022\" to access relevant content about Walmart's products from that year.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [144, 86]}<image>", "role": "user" }, { "content": "Thought: Click on the Walmart link to navigate directly to the store's TV selection.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [94, 461]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_4.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_5.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_6.png" ]
7,986
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Show me the best rated 4K TV on Walmart.<image>Thought: Click on \"walmart 2022\" to access relevant content about Walmart's products from that year.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [144, 86]}<image>Thought: Click on the Walmart link to navigate directly to the store's TV selection.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [94, 461]}<image>", "role": "user" }, { "content": "Thought: The task is complete as the GUI confirmed I am human by pressing and holding the button.\nAction: {\"action\": \"STATUS TASK COMPLETE\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/9085935698078452456_5.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_6.png", "training_data/AITW/aitw_images/webshopping/9085935698078452456_7.png" ]
7,987
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>", "role": "user" }, { "content": "Thought: Click on the Chrome app to open the browser for searching eBay.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [462, 1221]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_0.png" ]
7,988
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click on the Chrome app to open the browser for searching eBay.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [462, 1221]}<image>", "role": "user" }, { "content": "Thought: Click the \"Proceed to checkout\" button to finalize the purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [279, 106]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_0.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_1.png" ]
7,989
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click on the Chrome app to open the browser for searching eBay.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [462, 1221]}<image>Thought: Click the \"Proceed to checkout\" button to finalize the purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [279, 106]}<image>", "role": "user" }, { "content": "Thought: Type \"ebay.com\" in the search bar to navigate to eBay.\nAction: {\"action\": \"TYPE\", \"value\": \"ebay.com\", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_0.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_1.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_2.png" ]
7,990
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click the \"Proceed to checkout\" button to finalize the purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [279, 106]}<image>Thought: Type \"ebay.com\" in the search bar to navigate to eBay.\nAction: {\"action\": \"TYPE\", \"value\": \"ebay.com\", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click the link for ebay.com to navigate to the eBay website where I can manage my cart items.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [168, 208]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_1.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_2.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_3.png" ]
7,991
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Type \"ebay.com\" in the search bar to navigate to eBay.\nAction: {\"action\": \"TYPE\", \"value\": \"ebay.com\", \"position\": null}<image>Thought: Click the link for ebay.com to navigate to the eBay website where I can manage my cart items.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [168, 208]}<image>", "role": "user" }, { "content": "Thought: Click on the SHOPPING CART icon to access the items currently in the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [608, 176]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_2.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_3.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_4.png" ]
7,992
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click the link for ebay.com to navigate to the eBay website where I can manage my cart items.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [168, 208]}<image>Thought: Click on the SHOPPING CART icon to access the items currently in the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [608, 176]}<image>", "role": "user" }, { "content": "Thought: Click Remove to delete the item from the cart displayed on the shopping cart GUI.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [647, 730]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_3.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_4.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_5.png" ]
7,993
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click on the SHOPPING CART icon to access the items currently in the cart.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [608, 176]}<image>Thought: Click Remove to delete the item from the cart displayed on the shopping cart GUI.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [647, 730]}<image>", "role": "user" }, { "content": "Thought: Click the magnifying glass to access the search function on eBay.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [434, 170]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_4.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_5.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_6.png" ]
7,994
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click Remove to delete the item from the cart displayed on the shopping cart GUI.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [647, 730]}<image>Thought: Click the magnifying glass to access the search function on eBay.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [434, 170]}<image>", "role": "user" }, { "content": "Thought: Type \"bose soundlink mini\" in the search bar to find the desired product on eBay.\nAction: {\"action\": \"TYPE\", \"value\": \"bose soundlink mini \", \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_5.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_6.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_7.png" ]
7,995
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click the magnifying glass to access the search function on eBay.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [434, 170]}<image>Thought: Type \"bose soundlink mini\" in the search bar to find the desired product on eBay.\nAction: {\"action\": \"TYPE\", \"value\": \"bose soundlink mini \", \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click on the first entry of \"bose soundlink mini\" to view the product details.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [212, 462]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_6.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_7.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_8.png" ]
7,996
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Type \"bose soundlink mini\" in the search bar to find the desired product on eBay.\nAction: {\"action\": \"TYPE\", \"value\": \"bose soundlink mini \", \"position\": null}<image>Thought: Click on the first entry of \"bose soundlink mini\" to view the product details.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [212, 462]}<image>", "role": "user" }, { "content": "Thought: Click on the \"Bose SoundLink Mini Portable Bluetooth Speaker - Silver\" to view details about the item.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [294, 622]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_7.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_8.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_9.png" ]
7,997
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click on the first entry of \"bose soundlink mini\" to view the product details.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [212, 462]}<image>Thought: Click on the \"Bose SoundLink Mini Portable Bluetooth Speaker - Silver\" to view details about the item.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [294, 622]}<image>", "role": "user" }, { "content": "Thought: Scroll down to view more details about the Bose SoundLink Mini speaker.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_8.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_9.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_10.png" ]
7,998
[ { "content": "You are an assistant trained to navigate the mobile phone. \nGiven a task instruction, a screen observation, and an action history sequence, \noutput the next action and wait for the next observation. \nHere is the action space:\n1. `CLICK`: Click on an element, value is not applicable and the position [x,y] is required. \n2. `TYPE`: Type a string into an element, value is a string to type and the position is not applicable.\n3. `SELECT`: Select a value for an element, value is the value to select and the position is not applicable.\n4. `SCROLL UP`: Scroll up for the screen.\n5. `SCROLL DOWN`: Scroll down for the screen.\n6. `SCROLL LEFT`: Scroll left for the screen.\n7. `SCROLL RIGHT`: Scroll right for the screen.\n8. `PRESS BACK`: Press for returning to the previous step, value and position are not applicable.\n9. `PRESS HOME`: Press for returning to the home screen, value and position are not applicable.\n10. `PRESS ENTER`: Press for submitting the input content, value and position are not applicable.\n11. `STATUS TASK COMPLETE`: Indicate the task is completed, value and position are not applicable.\n12. `STATUS TASK IMPOSSIBLE `: Indicate the task is impossible to complete, value and position are not applicable.\n\nFormat your response as\nThought: <your reasoning process>\nAction: <the next action>\n\nFormat the action as a JSON object with the following keys:\n{\"action\": \"ACTION_TYPE\", \"value\": \"element\", \"position\": [x,y]}\n\nIf value or position is not applicable, set it as `None`.\nPosition represents the absolute pixel coordinates on the screenshot.\nTask: Clear all items from cart on ebay. Search for bose soundlink mini on ebay, select the first entry, add it to the cart, then select checkout.<image>Thought: Click on the \"Bose SoundLink Mini Portable Bluetooth Speaker - Silver\" to view details about the item.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [294, 622]}<image>Thought: Scroll down to view more details about the Bose SoundLink Mini speaker.\nAction: {\"action\": \"SCROLL DOWN\", \"value\": null, \"position\": null}<image>", "role": "user" }, { "content": "Thought: Click \"Add to cart\" to include the item in your shopping cart for future purchase.\nAction: {\"action\": \"CLICK\", \"value\": null, \"position\": [338, 1157]}", "role": "assistant" } ]
[ "training_data/AITW/aitw_images/webshopping/10428628259603550118_9.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_10.png", "training_data/AITW/aitw_images/webshopping/10428628259603550118_11.png" ]
7,999