dataset string | source_index int64 | source_json string | id string | image string | conversations list | entities list | serper_query string | search_queries list | references list |
|---|---|---|---|---|---|---|---|---|---|
GeoChat_Instruct | 0 | trainjson/GeoChat_Instruct_train.json | church/church_144.jpg | NWPU-RESISC45/church/church_144.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: railway station, harbor, desert, palace, mobile home park, cloud, storage tank, medium residential, ship, tennis court, golf course, baseball diamond, snowberg, river, mountain, church, terrace, bridge, ... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.438,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 1 | trainjson/GeoChat_Instruct_train.json | church/church_386.jpg | NWPU-RESISC45/church/church_386.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: palace, wetland, ground track field, church, intersection, cloud, river, harbor, lake, sparse residential, parking lot, tennis court, railway. \nAnswer in one word or a short phrase."
},
{
"from"... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.415,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 2 | trainjson/GeoChat_Instruct_train.json | church/church_362.jpg | NWPU-RESISC45/church/church_362.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: chaparral, cloud, tennis court, river, church, sparse residential, bridge, desert, beach, airport, roundabout, circular farmland, terrace, forest, rectangular farmland, commercial area, thermal power sta... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.422,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 3 | trainjson/GeoChat_Instruct_train.json | church/church_421.jpg | NWPU-RESISC45/church/church_421.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: ground track field, chaparral, harbor, circular farmland, desert, ship, railway station, meadow, bridge, mountain, forest, river, sparse residential, rectangular farmland, railway, freeway, overpass, com... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.612,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 4 | trainjson/GeoChat_Instruct_train.json | church/church_448.jpg | NWPU-RESISC45/church/church_448.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: golf course, mobile home park, runway, circular farmland, meadow, roundabout, bridge, harbor, freeway, sparse residential, thermal power station, overpass, church, industrial area, rectangular farmland, ... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.541,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 5 | trainjson/GeoChat_Instruct_train.json | church/church_600.jpg | NWPU-RESISC45/church/church_600.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: roundabout, river, baseball diamond, golf course, dense residential, church, commercial area. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.414,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 6 | trainjson/GeoChat_Instruct_train.json | church/church_176.jpg | NWPU-RESISC45/church/church_176.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: stadium, island, meadow, roundabout, runway, overpass, bridge, wetland, church, circular farmland, snowberg, terrace, thermal power station, medium residential. \nAnswer in one word or a short phrase."
... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.693,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 7 | trainjson/GeoChat_Instruct_train.json | church/church_216.jpg | NWPU-RESISC45/church/church_216.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: railway, church. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 3.301,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 8 | trainjson/GeoChat_Instruct_train.json | church/church_396.jpg | NWPU-RESISC45/church/church_396.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: wetland, railway station, mobile home park, thermal power station, terrace, airplane, lake, cloud, overpass, roundabout, mountain, palace, stadium, forest, ship, church, dense residential, runway, sparse... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 0.998,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 9 | trainjson/GeoChat_Instruct_train.json | church/church_289.jpg | NWPU-RESISC45/church/church_289.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: meadow, palace, wetland, parking lot, chaparral, roundabout, ship, lake, church, industrial area, commercial area, beach, airport, basketball court, stadium, railway, railway station, bridge, circular fa... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.986,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 10 | trainjson/GeoChat_Instruct_train.json | church/church_518.jpg | NWPU-RESISC45/church/church_518.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: church, snowberg, terrace. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church",
"terrace"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.917,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 11 | trainjson/GeoChat_Instruct_train.json | church/church_404.jpg | NWPU-RESISC45/church/church_404.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: golf course, baseball diamond, island, storage tank, railway station, mobile home park, thermal power station, beach, desert, cloud, forest, chaparral, stadium, tennis court, overpass, commercial area, m... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.92,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": "... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 12 | trainjson/GeoChat_Instruct_train.json | church/church_324.jpg | NWPU-RESISC45/church/church_324.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: bridge, commercial area, intersection, dense residential, overpass, ground track field, lake, sea ice, church, beach, sparse residential. \nAnswer in one word or a short phrase."
},
{
"from": "gp... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 3.037,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 13 | trainjson/GeoChat_Instruct_train.json | church/church_650.jpg | NWPU-RESISC45/church/church_650.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: circular farmland, parking lot, terrace, ship, overpass, chaparral, snowberg, industrial area, intersection, cloud, ground track field, dense residential, harbor, airplane, church, rectangular farmland, ... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.954,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 14 | trainjson/GeoChat_Instruct_train.json | church/church_184.jpg | NWPU-RESISC45/church/church_184.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: overpass, intersection, harbor, church, cloud, circular farmland, wetland, airplane, sparse residential, desert, roundabout. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"valu... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.883,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 15 | trainjson/GeoChat_Instruct_train.json | church/church_086.jpg | NWPU-RESISC45/church/church_086.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: palace, desert, ground track field, river, storage tank, sparse residential, bridge, airplane, church, industrial area, wetland, chaparral, mountain, commercial area, meadow, golf course, rectangular far... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.342,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 16 | trainjson/GeoChat_Instruct_train.json | church/church_410.jpg | NWPU-RESISC45/church/church_410.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: cloud, sea ice, storage tank, chaparral, river, dense residential, railway, freeway, commercial area, roundabout, mobile home park, forest, terrace, snowberg, sparse residential, circular farmland, railw... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.174,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 17 | trainjson/GeoChat_Instruct_train.json | church/church_128.jpg | NWPU-RESISC45/church/church_128.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: baseball diamond, tennis court, intersection, circular farmland, church, terrace, chaparral. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.447,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 18 | trainjson/GeoChat_Instruct_train.json | church/church_277.jpg | NWPU-RESISC45/church/church_277.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: railway station, church, railway, intersection. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.691,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 19 | trainjson/GeoChat_Instruct_train.json | church/church_193.jpg | NWPU-RESISC45/church/church_193.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: terrace, church, cloud. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.641,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 20 | trainjson/GeoChat_Instruct_train.json | church/church_383.jpg | NWPU-RESISC45/church/church_383.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: roundabout, church, ground track field, snowberg, railway, thermal power station, basketball court. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.474,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 21 | trainjson/GeoChat_Instruct_train.json | church/church_465.jpg | NWPU-RESISC45/church/church_465.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: runway, beach, parking lot, rectangular farmland, cloud, church. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.516,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 22 | trainjson/GeoChat_Instruct_train.json | church/church_455.jpg | NWPU-RESISC45/church/church_455.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: basketball court, snowberg, cloud, mobile home park, church, meadow, railway station, freeway, dense residential. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.282,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial View Church images - Shutterstock",
"time": "",
"url": "https://www.shutterstock.com/search/aerial-view-church",
"content": "458,161 aerial view church stock photos, vectors, and illustrations are available royalty-free for download.",
"rank": 1,
"serper_raw": {
"tit... |
GeoChat_Instruct | 23 | trainjson/GeoChat_Instruct_train.json | church/church_306.jpg | NWPU-RESISC45/church/church_306.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: commercial area, harbor, palace, ship, rectangular farmland, mobile home park, golf course, stadium, terrace, baseball diamond, medium residential, dense residential, desert, sea ice, beach, chaparral, a... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.542,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 24 | trainjson/GeoChat_Instruct_train.json | church/church_612.jpg | NWPU-RESISC45/church/church_612.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: tennis court, snowberg, meadow, bridge, runway, mountain, circular farmland, harbor, beach, thermal power station, sparse residential, mobile home park, airplane, golf course, island, overpass, storage t... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.859,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 25 | trainjson/GeoChat_Instruct_train.json | church/church_481.jpg | NWPU-RESISC45/church/church_481.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: cloud, dense residential, roundabout, church, thermal power station, island, baseball diamond, terrace, sparse residential, storage tank, bridge. \nAnswer in one word or a short phrase."
},
{
"fr... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 2.391,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 26 | trainjson/GeoChat_Instruct_train.json | church/church_328.jpg | NWPU-RESISC45/church/church_328.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: rectangular farmland, river, forest, sparse residential, snowberg, beach, airplane, lake, storage tank, harbor, terrace, church, circular farmland, bridge, dense residential, freeway, airport. \nAnswer i... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.242,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 27 | trainjson/GeoChat_Instruct_train.json | church/church_189.jpg | NWPU-RESISC45/church/church_189.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: intersection, church. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.45,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": "... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 28 | trainjson/GeoChat_Instruct_train.json | church/church_553.jpg | NWPU-RESISC45/church/church_553.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: parking lot, roundabout, dense residential, airplane, storage tank, mountain, runway, medium residential, golf course, overpass, railway station, palace, None, lake, meadow, forest. \nAnswer in one word ... | [
"None",
"church"
] | None remote sensing aerial image | [
{
"entity": "None",
"q": "None remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.787,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "None remote sensing aerial image",
"gl": "us",
... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 29 | trainjson/GeoChat_Instruct_train.json | church/church_492.jpg | NWPU-RESISC45/church/church_492.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: golf course, ground track field, basketball court, overpass, intersection, terrace, circular farmland, snowberg, rectangular farmland, commercial area, dense residential, roundabout, meadow, desert, ther... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.649,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 30 | trainjson/GeoChat_Instruct_train.json | church/church_377.jpg | NWPU-RESISC45/church/church_377.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: baseball diamond, bridge, beach, wetland, airplane, palace, forest, sparse residential, sea ice, dense residential, rectangular farmland, None, tennis court, mobile home park. \nAnswer in one word or a s... | [
"None",
"church"
] | None remote sensing aerial image | [
{
"entity": "None",
"q": "None remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.706,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "None remote sensing aerial image",
"gl": "us",
... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 31 | trainjson/GeoChat_Instruct_train.json | church/church_014.jpg | NWPU-RESISC45/church/church_014.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: commercial area, airport, dense residential, parking lot, basketball court, sea ice, stadium, tennis court, beach, roundabout, terrace, mountain, runway, desert, church, sparse residential, intersection,... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.225,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 32 | trainjson/GeoChat_Instruct_train.json | church/church_471.jpg | NWPU-RESISC45/church/church_471.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: terrace, church. \nAnswer in one word or a short phrase."
},
{
"from": "gpt",
"value": "church"
}
] | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.238,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
GeoChat_Instruct | 33 | trainjson/GeoChat_Instruct_train.json | church/church_562.jpg | NWPU-RESISC45/church/church_562.jpg | [
{
"from": "human",
"value": "<image>\nClassify the given image in one of the following classes. Classes: church, storage tank, overpass, parking lot, mobile home park, industrial area, palace, railway station, golf course, cloud, meadow, railway, lake, snowberg, stadium, chaparral, ground track field, comme... | [
"church"
] | church remote sensing aerial image | [
{
"entity": "church",
"q": "church remote sensing aerial image",
"returned": 5,
"added": 5,
"elapsed_seconds": 1.315,
"search_engine": "serper.dev",
"serper_endpoint": "https://google.serper.dev/search",
"serper_params": {
"q": "church remote sensing aerial image",
"gl": ... | [
{
"title": "Aerial Photographs and Satellite Images",
"time": "",
"url": "https://pubs.usgs.gov/gip/AerialPhotos_SatImages/aerial.html",
"content": "Aerial and satellite images, known as remotely sensed images, permit accurate mapping of land cover and make landscape features understandable.",
"... |
End of preview. Expand in Data Studio
KANT-8 Google RAG Corpus
This folder is an HF-ready text-only release of the KANT-8 training JSONs with Google/Serper search results embedded into each original sample. Images are not copied into this release; each record keeps the original image field as text.
Each row preserves the original training JSON fields and adds:
dataset: KANT dataset name.source_index: index in the originaltrainjson/*_train.jsonfile.source_json: local source JSON path used to build the row.entities: query-oriented entities used for search.serper_query: first Google/Serper query for the row, when available.search_queries: structured query metadata.references: up to five Google/Serper search references withtitle,time,url,content,rank,serper_raw,search_entity,search_query,search_engine,serper_endpoint, andserper_paramswhen available.
Fields that expose retrieval time or redundant source labeling, including retrieved_at, source_type, and content_extended, are intentionally removed.
Files
data/by_dataset/*.jsonl: one JSONL shard per KANT-8 dataset.manifest.json: generation metadata and record counts.
Summary
| Dataset | Records | Records with references | References |
|---|---|---|---|
GeoChat_Instruct |
308861 | 250941 | 1254703 |
llava_med |
40255 | 34536 | 172677 |
atom |
23941 | 23819 | 119095 |
art |
69812 | 69210 | 346046 |
astro |
29728 | 19666 | 98283 |
agri |
6160 | 6050 | 30250 |
chem |
66166 | 66166 | 330749 |
climate |
40000 | 39999 | 199995 |
Total records: 584923
Total references: 2551798
- Downloads last month
- 33