TanmayShah29/satquery-router-qwen2.5-1.5b-lora
Text Generation • Updated
Datasets:
id stringlengths 25 25 | query stringlengths 40 155 | hints dict | label dict | slices listlengths 1 1 | source stringclasses 1
value |
|---|---|---|---|---|---|
train-gpt-oss-120b-024-05 | clear sky, what is the sigma‑0 value for the runway surface in the SAR image? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-059-05 | Did the building’s SAR response differ between the pre‑event and the current image? | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": "Pre-Event -> Now",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-043-07 | Compare the flood water spread in the southern marsh from 2021-06-15 -> 2022-06-15, need SAR view cause optical cloudy. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [
"crossmodal_fusion"
],
"entities": [
"flood"
],
"temporal": "2021-06-15 -> 2022-06-15",
"cloud_piercing_required": true
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-014-06 | Describe the overall layout of the Kolkata harbour, focus on ships, warehouses and any green patches. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"building",
"maritime",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-082-07 | Has the number of cargo vessels anchored at the port changed since the image taken just before the monsoon arrived? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"maritime"
],
"temporal": "Pre-Event -> Post-Event",
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-021-03 | Provide a concise caption of the flood‑prone lowlands near the Yamuna in Delhi. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"flood"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-046-05 | Use SAR to look beneath the smoke at the excavation site and draw its perimeter | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"excavation"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-050-08 | Summarise the land‑cover and infrastructure near the Mumbai airport runway that touches the coastline. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"runway",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-003-00 | Please compare the tea plantation waterlogged areas before monsoon and after monsoon in the Assam estate. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"flood",
"vegetation"
],
"temporal": "Pre-Event -> Post-Event",
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-084-05 | mark the areas where the earthworks for the new relief road have started | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"excavation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-024-09 | clear weather, what does the cross‑polarisation (VH) tell us about the flood extent? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"flood"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-044-09 | Any signs of flood water receding on the southern side | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"flood"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-083-00 | Please highlight the main cargo vessel docked at the Mumbai port and outline nearby warehouses | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"building",
"maritime"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-008-02 | Is there any visible water pooling in the excavation trench south of the facility? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"flood",
"excavation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-024-00 | no clouds today, what's the VV backscatter level over the runway? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-067-00 | The monsoon haze is obscuring the runway; use SAR to reveal its condition. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-071-05 | Exclude the harbor, what is the SAR signal strength over the green belt along the river? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"flood",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-063-01 | What is the polarisation ratio over the launchpad, and can you highlight the launchpad area on the map? | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "sar_vqa",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"launchpad"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-002-02 | Pinpoint the cluster of residential buildings slated for redevelopment and summarize the surrounding land use. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-080-07 | I would like to know if the floodplain to the east of the runway is currently dry, ignoring any change the colour might suggest. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway",
"flood"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-079-01 | Is there any visible runway extension near the old airstrip pls? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-071-08 | No haze, is the SAR return from the drainage canal higher than from the surrounding fields? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"excavation",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-026-08 | Show the evolution of vegetation cover under the dust cloud from 2 Months Ago to Now and outline the newly greened areas | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [
"crossmodal_fusion",
"visual_grounding"
],
"entities": [
"vegetation"
],
"temporal": "2 Months Ago -> Now",
"cloud_piercing_required": true
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-008-09 | Is there any sign of a temporary storage building erected next to the runway extension? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway",
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-012-05 | What variations do you notice in the forest cover surrounding the launch pad when comparing today's view with the record from six months back? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"launchpad",
"vegetation"
],
"temporal": "6 Months Ago -> Now",
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-019-04 | Could you highlight any additional storage sheds that have been constructed at the dock between 2021 and 2023? | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": "2021 -> 2023",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-090-01 | kya is image mein runway ka surface wet lag raha hai? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-047-08 | Is there a visible boat docked at the smaller jetty located downstream of the main port? | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"maritime"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-028-06 | Do the satellite pictures show a boat moored at the nearby harbour? | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"maritime"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-005-03 | Describe the overall layout of the Bengaluru outskirt base and mark the location of the main building complex. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-053-08 | is there more waterlogged area after the heavy rain, before and after images needed | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"flood"
],
"temporal": "Pre-Event -> Post-Event",
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-104-02 | Describe your method for cloud storage fusion so I can see the runway extensions that are currently masked by fog | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-044-02 | Vegetation density ka count east side me batao | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-059-03 | Show whether the excavation site’s radar return has changed since the January 2023 capture. | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"excavation"
],
"temporal": "January 2023 -> Now",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-091-01 | Show me what changed at the launch pad between March 2023 and March 2024 and provide a brief scene summary. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"launchpad"
],
"temporal": "March 2023 -> March 2024",
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-089-01 | Clear sky, but the haze masks the launch pad; combine SAR and optical to see the propellant tanks. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [],
"entities": [
"launchpad"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-020-04 | Sky is clear, create a fused image to count vessels on the Brahmaputra, but ignore any ships docked at the jetty | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"maritime"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-031-08 | thick cloud layer above the plantation; provide a short scene caption after cloud removal | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"vegetation"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-086-03 | Show me the location of the vegetation patches and summarize the overall land cover. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "visual_grounding",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-102-02 | I would like to know how the vegetation near the river has altered since the monsoon, given that the atmosphere is clear now. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"flood",
"vegetation"
],
"temporal": "Monsoon -> Now",
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-034-08 | Change the colour scheme in your report and give a narrative caption of the entire Sriharikota site, focusing on vegetation and water features. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"flood",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-091-08 | What are the changes in vegetation cover between the dry season image and the monsoon image, and mark the regions where trees remain. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"vegetation"
],
"temporal": "Pre-Event -> Post-Event",
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-076-08 | Thick fog hides the agricultural fields near the city outskirts; could you pierce the fog and provide a brief scene caption? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"vegetation"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-063-04 | For the excavation trench, what is the VV backscatter level, and could you provide a short caption of the whole scene? | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "sar_vqa",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"excavation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-052-02 | Sky is clear, report the number of runways intersecting the mangrove area. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-015-04 | The sky is clear, not cloudy; what does the cross‑polarisation ratio tell us about the building rooftops in the industrial zone? | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-041-09 | is there any visible construction scaffolding near the administrative building? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-104-08 | Upload the data to cloud storage and fuse the images so I can see the building rooftops that the smog is masking over the central business district | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-042-04 | Describe your method for detecting the launchpad but tell me if the rocket tower is in view | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"launchpad"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-076-02 | Smoke from the recent fire hides the warehouse district; could you penetrate the haze and give a brief description of the scene? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-046-07 | Remove the thin cloud layer above the forest, fuse the bands and mark the clearings | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"vegetation"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-023-00 | visakhapatnam ke tarmac par kitne aircraft dikh rahe hain? | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-058-00 | Count the launchpads visible, noting there are no clouds in the sky | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"launchpad"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-104-05 | Draw conclusions from a fused SAR‑optical product to locate the narrow excavation trench that the dust storm is covering near the railway line | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"excavation"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-052-05 | No clouds, how many houses are still standing in the flood‑affected zone? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"building",
"flood"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-001-04 | Provide the count of mangrove trees and describe the overall land‑cover in this sector. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-005-08 | Count the number of helicopters on the tarmac and outline the parking spots they occupy. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-059-07 | Check if the vegetation’s radar backscatter has reduced since the post‑monsoon image. | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"vegetation"
],
"temporal": "Post-Event -> Now",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-009-08 | Identify the clusters of huts that are still intact after the inundation. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"building",
"flood"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-046-00 | Fuse SAR and optical to reveal the runway under haze and outline its edges | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-094-07 | No fog, use SAR to see beneath the tea canopy for any hidden flood patches. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [],
"entities": [
"flood",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-032-06 | What is the speckle noise level in the SAR image and box the launch pad structures present | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"launchpad"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-021-08 | Provide an overview of the runway complex at Indira Gandhi International Airport, including surrounding structures. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"runway",
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-059-02 | Is the vegetation backscatter higher now than in the pre‑event image near the hilltop? | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"vegetation"
],
"temporal": "Pre-Event -> Now",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-056-08 | describe the land cover, especially looking for any new building activity | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-017-04 | Is the tarmac clear of debris and can you mark any obstacles you see? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-019-09 | Please let me know if the green cover around the dock’s perimeter has decreased since the image captured in January 2023. | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"vegetation"
],
"temporal": "January 2023 -> Now",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-046-03 | Clouds obscure the river valley; fuse the data sources, outline the watercourse and summarize the surrounding vegetation | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding",
"scene_captioning"
],
"entities": [
"flood",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-073-01 | Cloud storage request: count the number of warehouses visible near the docks. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-022-02 | Show me the location of the fuel depot tanks close to the airstrip's western edge. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"runway",
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-012-09 | Please compare the shoreline contour at the dock area now with its appearance in the image taken three months prior. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"maritime"
],
"temporal": "3 Months Ago -> Now",
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-093-07 | I need a concise summary of the visual layout of the port’s western side, including any vegetation. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"maritime",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-016-01 | outline the expanded berth area that’s been added since january 2021 | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"maritime"
],
"temporal": "January 2021 -> Now",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-024-03 | excluding the ships, what does the VH backscatter indicate about the trench soil? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"excavation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-091-06 | Identify any new runway cracks compared to the image from 2 Years Ago and outline them on the map. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"runway"
],
"temporal": "2 Years Ago -> Now",
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-105-02 | What differences are there in the launch pad structures from 1 Year Ago to Now? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"launchpad"
],
"temporal": "1 Year Ago -> Now",
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-048-00 | Please change the colour of the runway markings and show me exactly where the new taxiway starts. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-048-06 | Mark as done the location of the abandoned building near the riverbank and outline it. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"building",
"flood"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-083-08 | kya yahan koi vegetation ka cluster hai jo flood ke baad survive kar raha hai, point out karo | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"flood",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-089-03 | Exclude the farmland, focus on the construction site; fuse the data to map any new buildings under the thin cloud. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-073-03 | Draw conclusions about the runway length and say if any aircraft are parked. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-006-01 | No clouds over the launch complex, just describe the buildings, launchpad structures and surrounding forest in detail. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"launchpad",
"building",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-069-09 | Give an overview of the area and pinpoint any ships docked along the shore. | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "scene_captioning",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"maritime"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-034-01 | Please store the cloud data and provide a detailed scene caption covering the launchpad facilities and nearby fields. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"launchpad",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-080-00 | Could you change the colour of the runway in the image so I can see cracks better? | {
"has_bitemporal_pair": false,
"has_sar_pair": false
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-010-08 | Generate a fused view to locate any flood water trapped in low‑lying fields under haze. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [],
"entities": [
"flood",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-100-07 | indicate where the runway taxiway intersects the service road | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-026-05 | Assess the change in excavation depth from pre‑event to post‑event and highlight the expanded trench network | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"excavation"
],
"temporal": "Pre-Event -> Post-Event",
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-001-00 | How many fishing vessels are docked and outline the nearby vegetation patches? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"maritime",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-104-04 | Mark as done the SAR‑optical fusion and highlight any launchpad structures concealed by the heavy haze over the coastal area | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "crossmodal_fusion",
"secondary_intents": [
"visual_grounding"
],
"entities": [
"launchpad"
],
"temporal": null,
"cloud_piercing_required": true
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-032-07 | Provide the polarimetric decomposition result and give a brief caption of the vegetation and water mix in the area | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [
"scene_captioning"
],
"entities": [
"flood",
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"compound"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-056-04 | rann ke north side ka scene describe karo, koi ship ya boat to nahi hai na? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "scene_captioning",
"secondary_intents": [],
"entities": [
"maritime"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"typo_hinglish"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-042-02 | Point cloud the runway area and report if any tarmac cracks are visible | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"keyword_trap"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-058-05 | How many excavation sites are visible, with clear skies and no smoke | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"excavation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-084-04 | identify the locations of the fishing boats that are still anchored after the storm surge | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"maritime"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-100-02 | mark the building complex near the north edge of the scene | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "visual_grounding",
"secondary_intents": [],
"entities": [
"building"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-029-05 | Provide the backscatter coefficient values for the airport runway at Sardar Vallabhbhai Patel International Airport. | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "sar_vqa",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"clean"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-059-00 | Did the backscatter intensity over the Visakhapatnam river increase compared to the image from two months ago? | {
"has_bitemporal_pair": true,
"has_sar_pair": false
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"flood"
],
"temporal": "2 Months Ago -> Now",
"cloud_piercing_required": false
} | [
"hint_driven"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-041-08 | are the tree lines intact along the northern perimeter? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "optical_vqa",
"secondary_intents": [],
"entities": [
"vegetation"
],
"temporal": null,
"cloud_piercing_required": false
} | [
"paraphrase"
] | openai/gpt-oss-120b |
train-gpt-oss-120b-040-06 | The area is cloud‑free; are there any new runway markings visible compared with the pre‑event image? | {
"has_bitemporal_pair": false,
"has_sar_pair": true
} | {
"intent": "bitemporal_change",
"secondary_intents": [],
"entities": [
"runway"
],
"temporal": "Pre-Event -> Post-Event",
"cloud_piercing_required": false
} | [
"negation"
] | openai/gpt-oss-120b |
End of preview. Expand in Data Studio
satquery-router-decisions
Synthetic satellite-imagery questions, each with the two UI hints the SatQuery app sends and a routing decision (schema v1.0). Used to train and evaluate TanmayShah29/satquery-router-qwen2.5-1.5b-lora. Pipeline, labelling guidelines and full data card: https://github.com/TanmayShah29/llm-query-router
from datasets import load_dataset
ds = load_dataset("TanmayShah29/satquery-router-decisions")
Splits
| split | rows | queries written by | labels |
|---|---|---|---|
| train | 564 | openai/gpt-oss-120b |
the generator's label, kept only if nvidia/nemotron-3-ultra-550b-a55b produced the identical label blind |
| validation | 93 | same pool as train | same |
| test | 189 | qwen/qwen3.8-27b, with its own personas and regions and no examples, before any training data existed |
kept only if qwen/qwen3.8-27b and openai/gpt-oss-20b labelled it identically, blind (intent kappa 0.889, 63% kept) |
| probes | 20 | hand-picked cases where the regex router fails | hand-labelled; a diagnostic, not a measurement |
All labels except the probes are silver: no human wrote or checked them. The test filter keeps only queries
two models found unambiguous, so it's easier than real traffic; only 35% of bitemporal-mode test candidates
survived it. Near-duplicates across splits were removed (character 3–5-gram TF-IDF cosine ≥ 0.85; none found).
The test split's sha256 at freeze time is recorded in the GitHub repo (data/router_ft/test.sha256, for the
original file with provenance columns).
Fields
id,queryhints:has_bitemporal_pair,has_sar_pair(UI mode; bitemporal mode means the intent is alwaysbitemporal_change; the SAR hint never changes the label)label:intent,secondary_intents,entities,temporal(null or"<start> -> <end>"),cloud_piercing_requiredslices: clean, paraphrase, negation, typo_hinglish, compound, hint_driven, keyword_trapsource: the model that wrote the query
Licence and limitations
Rows were written and labelled by gpt-oss-120b, gpt-oss-20b, qwen3.8-27b and Nemotron Ultra through
free API tiers; check those models' output terms before commercial reuse. Machine-written English and
Hinglish (Latin script); Indian place names dominate; hint distribution set by hand to match the app's
default mode. If you train on this data, don't evaluate on its test split.
- Downloads last month
- 42
Models trained or fine-tuned on TanmayShah29/satquery-router-decisions