Datasets:
question_id string | scene_id string | level string | subcategory string | question string | answer string | answer_type string | choices list | image_t1 string | image_t2 string | scene_graph string | record_json string | support_json string | program_json string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
RSF-Q00000001 | RSF-S00000001 | Perception | object_presence | Does the image contain any warehouse? | No | text | [
"Yes",
"Can't judge",
"No"
] | Perception/object_presence/images/RSF-S00000001_t1.jpg | null | Perception/object_presence/scene_graphs/RSF-S00000001.json | Perception/object_presence/object_presence.json | [{"claim_type":"Existence","time":"t1","subject":"warehouse","value":false,"refs":{"node_ids":[],"edge_ids":[]}}] | {"program_id":"perception.object_presence","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"warehouse"},"outputs":{"answer":"value"},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true}],"answer_fn":{"op":"bool_label","claim_type":"Existence","field":"value","var"... |
RSF-Q00000002 | RSF-S00000002 | Perception | object_counting | Count the aircraft shown in the image. | 7 | text | [
"6",
"7",
"8",
"9"
] | Perception/object_counting/images/RSF-S00000002_t1.jpg | null | Perception/object_counting/scene_graphs/RSF-S00000002.json | Perception/object_counting/object_counting.json | [{"claim_type":"Existence","time":"t1","subject":"aircraft","value":true,"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003","n_obj_00004","n_obj_00005","n_obj_00006","n_obj_00007"],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"aircraft","value":7,"refs":{"node_ids":["n_obj_00001","n_obj_000... | {"program_id":"perception.object_count","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"aircraft","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003","n_obj_00004","n_obj_00005","n_obj_00006","n_obj_00007"],"edge_ids":[]},"sr_requ... |
RSF-Q00000007 | RSF-S00000007 | Perception | object_localization | Is the vehicle located in the bottom-left region? | No | text | [
"Yes",
"Can't judge",
"No"
] | Perception/object_localization/images/RSF-S00000007_t1.jpg | null | Perception/object_localization/scene_graphs/RSF-S00000007.json | Perception/object_localization/object_localization.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center-left","refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}}] | {"program_id":"perception.object_location","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Location","scope":{"time":"t1"},"cons... |
RSF-Q00001290 | RSF-S00001290 | Perception | fine_grained_recognition | Which of the given types describes the large vehicle? | cargo truck | text | [
"cargo truck",
"truck tractor",
"trailer",
"dump truck"
] | Perception/fine_grained_recognition/images/RSF-S00001290_t1.tif | null | Perception/fine_grained_recognition/scene_graphs/RSF-S00001290.json | Perception/fine_grained_recognition/fine_grained_recognition.json | [{"claim_type":"Existence","time":"t1","subject":"large_vehicle","value":true,"refs":{"node_ids":["n_obj_00044"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"large_vehicle","name":"subtype","value":"cargo_truck","refs":{"node_ids":["n_obj_00044"],"edge_ids":[]}}] | {"program_id":"perception.object_attribute","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"large_vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00044"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1... |
RSF-Q00004581 | RSF-S00004581 | Relational reasoning | aggregate_distribution | Choose the area that best matches the main cluster of harbors. | mainly near the center of the image | text | [
"mainly near the center of the image",
"mainly in the lower-left part of the image",
"mainly in the center-right part of the image",
"mainly in the center-left part of the image"
] | Relational reasoning/aggregate_distribution/images/RSF-S00004581_t1.jpg | null | Relational reasoning/aggregate_distribution/scene_graphs/RSF-S00004581.json | Relational reasoning/aggregate_distribution/aggregate_distribution.json | [{"claim_type":"Counting","time":"t1","subject":"harbor::top-left","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"harbor::top-center","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"harbor::top-right","value":0,"refs":{"nod... | {"program_id":"spatial.aggregate_distribution","slots":[{"slot_id":"s_001","claim_type":"Counting","scope":{"time":"t1"},"constraints":{"subject":"harbor::top-left","quantity":0},"outputs":{},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Counting","scope":{"time":"t1"},"const... |
RSF-Q00004589 | RSF-S00004589 | Relational reasoning | directional | What listed object is to the right of the bridge? | the vehicle in the center-right | text | [
"the vehicle in the center-left",
"the vehicle in the center",
"the vehicle in the bottom-right",
"the vehicle in the center-right"
] | Relational reasoning/directional/images/RSF-S00004589_t1.jpg | null | Relational reasoning/directional/scene_graphs/RSF-S00004589.json | Relational reasoning/directional/directional.json | [{"claim_type":"Existence","time":"t1","subject":"bridge","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center-rig... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"bridge","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t1"... |
RSF-Q00004602 | RSF-S00004601 | Relational reasoning | proximity | Relative to the vehicle, which option is farthest away? | the harbor in the top-left | text | [
"the ship in the center-left",
"the harbor in the top-center",
"the bridge",
"the harbor in the top-left"
] | Relational reasoning/proximity/images/RSF-S00004601_t1.jpg | null | Relational reasoning/proximity/scene_graphs/RSF-S00004601.json | Relational reasoning/proximity/proximity.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00025"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"vehicle","name":"center","value":[738.0,655.5],"refs":{"node_ids":["n_obj_00025"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"s... | {"program_id":"spatial.proximity","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00025"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"constraints"... |
RSF-Q00004604 | RSF-S00004602 | Relational reasoning | topological | Which listed object is contained within the outline of the overpass? | the vehicle in the center-left | text | [
"the vehicle in the bottom-center",
"the vehicle in the bottom-right",
"the vehicle in the center-left",
"the vehicle in the top-left"
] | Relational reasoning/topological/images/RSF-S00004602_t1.jpg | null | Relational reasoning/topological/scene_graphs/RSF-S00004602.json | Relational reasoning/topological/topological.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center-l... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t... |
RSF-Q00004672 | RSF-S00004662 | Relational reasoning | projective_ordering | Choose the left-to-right order of the listed aircraft. | the A220, the Boeing 737, the A330 | text | [
"the A330, the Boeing 737, the A220",
"the Boeing 737, the A220, the A330",
"the Boeing 737, the A330, the A220",
"the A220, the Boeing 737, the A330"
] | Relational reasoning/projective_ordering/images/RSF-S00004662_t1.tif | null | Relational reasoning/projective_ordering/scene_graphs/RSF-S00004662.json | Relational reasoning/projective_ordering/projective_ordering.json | [{"claim_type":"Existence","time":"t1","subject":"a220","value":true,"refs":{"node_ids":["n_obj_00007"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"a220","name":"center","value":[57.0,154.5],"refs":{"node_ids":["n_obj_00007"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"boeing73... | {"program_id":"spatial.projective_ordering","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"a220","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00007"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"const... |
RSF-Q00009465 | RSF-S00008725 | Temporal reasoning | category_turnover | Which category appears only after the earlier image? | building | text | [
"building",
"low vegetation",
"non vegetated surface",
"tree"
] | Temporal reasoning/category_turnover/images/RSF-S00008725_t1.png | Temporal reasoning/category_turnover/images/RSF-S00008725_t2.png | Temporal reasoning/category_turnover/scene_graphs/RSF-S00008725.json | Temporal reasoning/category_turnover/category_turnover.json | [{"claim_type":"Existence","time":"t1","subject":"building","value":false,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","subject":"building","value":true,"refs":{"node_ids":["n_t2_buildings"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t2","subject":"building","name":"aggregate_area"... | {"program_id":"temporal.category_appearance","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"building"},"outputs":{"exists:building:t1":"value"},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t2"}... |
RSF-Q00009466 | RSF-S00008726 | Temporal reasoning | semantic_transition | From the earlier image to the later image, low vegetation mainly comes from which earlier category? | building | text | [
"non vegetated surface",
"tree",
"building",
"water"
] | Temporal reasoning/semantic_transition/images/RSF-S00008726_t1.png | Temporal reasoning/semantic_transition/images/RSF-S00008726_t2.png | Temporal reasoning/semantic_transition/scene_graphs/RSF-S00008726.json | Temporal reasoning/semantic_transition/semantic_transition.json | [{"claim_type":"Existence","time":"t1","subject":"building","value":true,"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"building","name":"aggregate_area","value":7315,"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","subj... | {"program_id":"temporal.semantic_transition_source","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"building","value":true},"outputs":{},"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"tim... |
RSF-Q00009468 | RSF-S00008727 | Temporal reasoning | net_change | Which category shows the largest increase in area between the two images? | tree | text | [
"non vegetated surface",
"water",
"tree",
"building"
] | Temporal reasoning/net_change/images/RSF-S00008727_t1.png | Temporal reasoning/net_change/images/RSF-S00008727_t2.png | Temporal reasoning/net_change/scene_graphs/RSF-S00008727.json | Temporal reasoning/net_change/net_change.json | [{"claim_type":"Existence","time":"t1","subject":"non_vegetated_surface","value":true,"refs":{"node_ids":["n_t1_NVG_surface"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"non_vegetated_surface","name":"aggregate_area","value":102076,"refs":{"node_ids":["n_t1_NVG_surface"],"edge_ids":[]}},{"claim_typ... | {"program_id":"temporal.max_area_increase","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"non_vegetated_surface"},"outputs":{"exists:non_vegetated_surface:t1":"value"},"refs":{"node_ids":["n_t1_NVG_surface"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","cla... |
RSF-Q00000004 | RSF-S00000004 | Perception | object_presence | Does the image contain any ship? | Yes | text | [
"Can't judge",
"Yes",
"No"
] | Perception/object_presence/images/RSF-S00000004_t1.jpg | null | Perception/object_presence/scene_graphs/RSF-S00000004.json | Perception/object_presence/object_presence.json | [{"claim_type":"Existence","time":"t1","subject":"ship","value":true,"refs":{"node_ids":["n_obj_00001","n_obj_00002"],"edge_ids":[]}}] | {"program_id":"perception.object_presence","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"ship"},"outputs":{"answer":"value"},"refs":{"node_ids":["n_obj_00001","n_obj_00002"],"edge_ids":[]},"sr_required":true}],"answer_fn":{"op":"bool_label","claim_type":"Existence"... |
RSF-Q00000003 | RSF-S00000003 | Perception | object_counting | Count the harbors shown in the image. | 3 | text | [
"2",
"3",
"4",
"5"
] | Perception/object_counting/images/RSF-S00000003_t1.jpg | null | Perception/object_counting/scene_graphs/RSF-S00000003.json | Perception/object_counting/object_counting.json | [{"claim_type":"Existence","time":"t1","subject":"harbor","value":true,"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003"],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"harbor","value":3,"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003"],"edge_ids":[]}}] | {"program_id":"perception.object_count","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"harbor","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Counting","sc... |
RSF-Q00000008 | RSF-S00000008 | Perception | object_localization | Is the vehicle located in the bottom-center region? | Yes | text | [
"No",
"Can't judge",
"Yes"
] | Perception/object_localization/images/RSF-S00000008_t1.jpg | null | Perception/object_localization/scene_graphs/RSF-S00000008.json | Perception/object_localization/object_localization.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00008"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"bottom-center","refs":{"node_ids":["n_obj_00008"],"edge_ids":[]}}] | {"program_id":"perception.object_location","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00008"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Location","scope":{"time":"t1"},"cons... |
RSF-Q00001293 | RSF-S00001293 | Perception | fine_grained_recognition | Which subtype best matches the ship in the center-left? | other ship | text | [
"dry cargo ship",
"engineering ship",
"fishing boat",
"other ship"
] | Perception/fine_grained_recognition/images/RSF-S00001293_t1.tif | null | Perception/fine_grained_recognition/scene_graphs/RSF-S00001293.json | Perception/fine_grained_recognition/fine_grained_recognition.json | [{"claim_type":"Existence","time":"t1","subject":"ship","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"ship","value":"center-left","refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"ship","name":"subtype","... | {"program_id":"perception.object_attribute","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"ship","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Location","scope":{"time":"t1"},"constr... |
RSF-Q00004582 | RSF-S00004582 | Relational reasoning | aggregate_distribution | Where do the tennis courts cluster most strongly? | mainly in the center-left part of the image | text | [
"mainly in the lower central part of the image",
"mainly in the center-left part of the image",
"mainly in the lower-left part of the image",
"mainly in the upper-right part of the image"
] | Relational reasoning/aggregate_distribution/images/RSF-S00004582_t1.png | null | Relational reasoning/aggregate_distribution/scene_graphs/RSF-S00004582.json | Relational reasoning/aggregate_distribution/aggregate_distribution.json | [{"claim_type":"Counting","time":"t1","subject":"tennis_court::top-left","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"tennis_court::top-center","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"tennis_court::top-right","val... | {"program_id":"spatial.aggregate_distribution","slots":[{"slot_id":"s_001","claim_type":"Counting","scope":{"time":"t1"},"constraints":{"subject":"tennis_court::top-left","quantity":0},"outputs":{},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Counting","scope":{"time":"t1"},... |
RSF-Q00004591 | RSF-S00004591 | Relational reasoning | directional | What listed object is to the right of the overpass? | the vehicle in the bottom-right | text | [
"the vehicle in the center-left",
"the vehicle in the bottom-right",
"the vehicle in the center",
"the vehicle in the bottom-left"
] | Relational reasoning/directional/images/RSF-S00004591_t1.jpg | null | Relational reasoning/directional/scene_graphs/RSF-S00004591.json | Relational reasoning/directional/directional.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"bottom-r... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t... |
RSF-Q00004655 | RSF-S00004648 | Relational reasoning | proximity | Which object sits nearest to the overpass? | the service area in the center | text | [
"the vehicle in the top-center",
"the service area in the center",
"the service area in the center-right",
"the vehicle in the bottom-right"
] | Relational reasoning/proximity/images/RSF-S00004648_t1.jpg | null | Relational reasoning/proximity/scene_graphs/RSF-S00004648.json | Relational reasoning/proximity/proximity.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00015"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"overpass","name":"center","value":[439.0,441.5],"refs":{"node_ids":["n_obj_00015"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":... | {"program_id":"spatial.proximity","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00015"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"constraints... |
RSF-Q00004620 | RSF-S00004616 | Relational reasoning | topological | Which listed object lies within the outline of the overpass? | the vehicle in the bottom-left | text | [
"the vehicle in the center-right",
"the vehicle in the center",
"the vehicle in the center-left",
"the vehicle in the bottom-left"
] | Relational reasoning/topological/images/RSF-S00004616_t1.jpg | null | Relational reasoning/topological/scene_graphs/RSF-S00004616.json | Relational reasoning/topological/topological.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00002"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"bottom-l... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t... |
RSF-Q00004745 | RSF-S00004730 | Relational reasoning | projective_ordering | Which sequence matches the listed aircraft when ordered left to right? | the A321, the A220, the A350 | text | [
"the A350, the A321, the A220",
"the A220, the A321, the A350",
"the A350, the A220, the A321",
"the A321, the A220, the A350"
] | Relational reasoning/projective_ordering/images/RSF-S00004730_t1.tif | null | Relational reasoning/projective_ordering/scene_graphs/RSF-S00004730.json | Relational reasoning/projective_ordering/projective_ordering.json | [{"claim_type":"Existence","time":"t1","subject":"a321","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"a321","name":"center","value":[404.5,246.5],"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"a220","... | {"program_id":"spatial.projective_ordering","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"a321","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"const... |
RSF-Q00009471 | RSF-S00008728 | Temporal reasoning | category_turnover | Which category appears in the earlier image but not in the later image? | playground | text | [
"non vegetated surface",
"low vegetation",
"building",
"playground"
] | Temporal reasoning/category_turnover/images/RSF-S00008728_t1.png | Temporal reasoning/category_turnover/images/RSF-S00008728_t2.png | Temporal reasoning/category_turnover/scene_graphs/RSF-S00008728.json | Temporal reasoning/category_turnover/category_turnover.json | [{"claim_type":"Existence","time":"t1","subject":"playground","value":true,"refs":{"node_ids":["n_t1_playgrounds"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"playground","name":"aggregate_area","value":21261,"refs":{"node_ids":["n_t1_playgrounds"],"edge_ids":[]}},{"claim_type":"Existence","time":"... | {"program_id":"temporal.category_disappearance","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"playground"},"outputs":{"exists:playground:t1":"value"},"refs":{"node_ids":["n_t1_playgrounds"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribu... |
RSF-Q00009467 | RSF-S00008726 | Temporal reasoning | semantic_transition | In the later image, what does the earlier building area mainly transition into? | low vegetation | text | [
"low vegetation",
"tree",
"non vegetated surface",
"water"
] | Temporal reasoning/semantic_transition/images/RSF-S00008726_t1.png | Temporal reasoning/semantic_transition/images/RSF-S00008726_t2.png | Temporal reasoning/semantic_transition/scene_graphs/RSF-S00008726.json | Temporal reasoning/semantic_transition/semantic_transition.json | [{"claim_type":"Existence","time":"t1","subject":"building","value":true,"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"building","name":"aggregate_area","value":7315,"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","subj... | {"program_id":"temporal.semantic_transition_target","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"building","value":true},"outputs":{},"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"tim... |
RSF-Q00009472 | RSF-S00008728 | Temporal reasoning | net_change | Which category undergoes the greatest decrease in area between the two images? | playground | text | [
"water",
"non vegetated surface",
"tree",
"playground"
] | Temporal reasoning/net_change/images/RSF-S00008728_t1.png | Temporal reasoning/net_change/images/RSF-S00008728_t2.png | Temporal reasoning/net_change/scene_graphs/RSF-S00008728.json | Temporal reasoning/net_change/net_change.json | [{"claim_type":"Existence","time":"t1","subject":"water","value":true,"refs":{"node_ids":["n_t1_water"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"water","name":"aggregate_area","value":319,"refs":{"node_ids":["n_t1_water"],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","subject":"water","v... | {"program_id":"temporal.max_area_decrease","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"water"},"outputs":{"exists:water:t1":"value"},"refs":{"node_ids":["n_t1_water"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"... |
RSF-Q00000005 | RSF-S00000005 | Perception | object_presence | Does the image contain any warehouse? | No | text | [
"No",
"Yes",
"Can't judge"
] | Perception/object_presence/images/RSF-S00000005_t1.jpg | null | Perception/object_presence/scene_graphs/RSF-S00000005.json | Perception/object_presence/object_presence.json | [{"claim_type":"Existence","time":"t1","subject":"warehouse","value":false,"refs":{"node_ids":[],"edge_ids":[]}}] | {"program_id":"perception.object_presence","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"warehouse"},"outputs":{"answer":"value"},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true}],"answer_fn":{"op":"bool_label","claim_type":"Existence","field":"value","var"... |
RSF-Q00000010 | RSF-S00000010 | Perception | object_counting | Count the vehicles shown in the image. | 7 | text | [
"9",
"8",
"6",
"7"
] | Perception/object_counting/images/RSF-S00000010_t1.jpg | null | Perception/object_counting/scene_graphs/RSF-S00000010.json | Perception/object_counting/object_counting.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00004","n_obj_00005","n_obj_00006","n_obj_00007","n_obj_00008","n_obj_00009","n_obj_00010"],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"vehicle","value":7,"refs":{"node_ids":["n_obj_00004","n_obj_00005... | {"program_id":"perception.object_count","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00004","n_obj_00005","n_obj_00006","n_obj_00007","n_obj_00008","n_obj_00009","n_obj_00010"],"edge_ids":[]},"sr_requi... |
RSF-Q00000009 | RSF-S00000009 | Perception | object_localization | Which region of the image contains the vehicle? | center | text | [
"top-right",
"top-center",
"top-left",
"center"
] | Perception/object_localization/images/RSF-S00000009_t1.jpg | null | Perception/object_localization/scene_graphs/RSF-S00000009.json | Perception/object_localization/object_localization.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00002"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center","refs":{"node_ids":["n_obj_00002"],"edge_ids":[]}}] | {"program_id":"perception.object_location","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00002"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Location","scope":{"time":"t1"},"cons... |
RSF-Q00001295 | RSF-S00001295 | Perception | fine_grained_recognition | Which subtype best matches the large vehicle? | dump truck | text | [
"cargo truck",
"trailer",
"truck tractor",
"dump truck"
] | Perception/fine_grained_recognition/images/RSF-S00001295_t1.tif | null | Perception/fine_grained_recognition/scene_graphs/RSF-S00001295.json | Perception/fine_grained_recognition/fine_grained_recognition.json | [{"claim_type":"Existence","time":"t1","subject":"large_vehicle","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"large_vehicle","name":"subtype","value":"dump_truck","refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}}] | {"program_id":"perception.object_attribute","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"large_vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1... |
RSF-Q00004583 | RSF-S00004583 | Relational reasoning | aggregate_distribution | Which description best matches the overall distribution of the train carriages? | split between two separated parts of the image | text | [
"clustered in one small part of the image",
"split between two separated parts of the image",
"arranged in a narrow horizontal band",
"mostly concentrated near the center"
] | Relational reasoning/aggregate_distribution/images/RSF-S00004583_t1.png | null | Relational reasoning/aggregate_distribution/scene_graphs/RSF-S00004583.json | Relational reasoning/aggregate_distribution/aggregate_distribution.json | [{"claim_type":"Counting","time":"t1","subject":"train_carriage::top-left","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"train_carriage::top-center","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"train_carriage::top-right... | {"program_id":"spatial.aggregate_distribution","slots":[{"slot_id":"s_001","claim_type":"Counting","scope":{"time":"t1"},"constraints":{"subject":"train_carriage::top-left","quantity":0},"outputs":{},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Counting","scope":{"time":"t1"... |
RSF-Q00004593 | RSF-S00004593 | Relational reasoning | directional | Which listed object is located to the left of the overpass? | the vehicle in the center-left | text | [
"the vehicle in the bottom-center",
"the vehicle in the center-left",
"the vehicle in the top-center",
"the vehicle in the center"
] | Relational reasoning/directional/images/RSF-S00004593_t1.jpg | null | Relational reasoning/directional/scene_graphs/RSF-S00004593.json | Relational reasoning/directional/directional.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00004"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center-l... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t... |
RSF-Q00004664 | RSF-S00004657 | Relational reasoning | proximity | Relative to the toll station, which option is farthest away? | the vehicle in the bottom-right | text | [
"the vehicle in the bottom-right",
"the vehicle in the top-center",
"the vehicle in the bottom-center",
"the vehicle in the center-right"
] | Relational reasoning/proximity/images/RSF-S00004657_t1.jpg | null | Relational reasoning/proximity/scene_graphs/RSF-S00004657.json | Relational reasoning/proximity/proximity.json | [{"claim_type":"Existence","time":"t1","subject":"toll_station","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"toll_station","name":"center","value":[484.0,431.5],"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","s... | {"program_id":"spatial.proximity","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"toll_station","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"constra... |
RSF-Q00004625 | RSF-S00004620 | Relational reasoning | topological | Which listed object falls inside the footprint of the overpass? | the vehicle in the center | text | [
"the vehicle in the bottom-left",
"the vehicle in the center-right",
"the vehicle in the top-left",
"the vehicle in the center"
] | Relational reasoning/topological/images/RSF-S00004620_t1.jpg | null | Relational reasoning/topological/scene_graphs/RSF-S00004620.json | Relational reasoning/topological/topological.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center",... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t... |
RSF-Q00004803 | RSF-S00004775 | Relational reasoning | projective_ordering | Which option correctly orders the listed aircraft from left to right? | the A350, the A321, the A220 | text | [
"the A350, the A220, the A321",
"the A350, the A321, the A220",
"the A220, the A321, the A350",
"the A321, the A220, the A350"
] | Relational reasoning/projective_ordering/images/RSF-S00004775_t1.tif | null | Relational reasoning/projective_ordering/scene_graphs/RSF-S00004775.json | Relational reasoning/projective_ordering/projective_ordering.json | [{"claim_type":"Existence","time":"t1","subject":"a350","value":true,"refs":{"node_ids":["n_obj_00004"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"a350","name":"center","value":[313.0,244.5],"refs":{"node_ids":["n_obj_00004"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"a321","... | {"program_id":"spatial.projective_ordering","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"a350","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00004"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"const... |
RSF-Q00009478 | RSF-S00008731 | Temporal reasoning | category_turnover | Which category is absent from the later image? | water | text | [
"low vegetation",
"water",
"building",
"non vegetated surface"
] | Temporal reasoning/category_turnover/images/RSF-S00008731_t1.png | Temporal reasoning/category_turnover/images/RSF-S00008731_t2.png | Temporal reasoning/category_turnover/scene_graphs/RSF-S00008731.json | Temporal reasoning/category_turnover/category_turnover.json | [{"claim_type":"Existence","time":"t1","subject":"water","value":true,"refs":{"node_ids":["n_t1_water"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"water","name":"aggregate_area","value":12718,"refs":{"node_ids":["n_t1_water"],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","subject":"water",... | {"program_id":"temporal.category_disappearance","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"water"},"outputs":{"exists:water:t1":"value"},"refs":{"node_ids":["n_t1_water"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"ti... |
RSF-Q00009469 | RSF-S00008727 | Temporal reasoning | semantic_transition | In the earlier image, what is the main source category of the later low vegetation area? | non vegetated surface | text | [
"non vegetated surface",
"tree",
"water",
"building"
] | Temporal reasoning/semantic_transition/images/RSF-S00008727_t1.png | Temporal reasoning/semantic_transition/images/RSF-S00008727_t2.png | Temporal reasoning/semantic_transition/scene_graphs/RSF-S00008727.json | Temporal reasoning/semantic_transition/semantic_transition.json | [{"claim_type":"Existence","time":"t1","subject":"non_vegetated_surface","value":true,"refs":{"node_ids":["n_t1_NVG_surface"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"non_vegetated_surface","name":"aggregate_area","value":102076,"refs":{"node_ids":["n_t1_NVG_surface"],"edge_ids":[]}},{"claim_typ... | {"program_id":"temporal.semantic_transition_source","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"non_vegetated_surface","value":true},"outputs":{},"refs":{"node_ids":["n_t1_NVG_surface"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute... |
RSF-Q00009474 | RSF-S00008729 | Temporal reasoning | net_change | Which category has the largest area gain across the two images? | tree | text | [
"low vegetation",
"tree",
"playground",
"building"
] | Temporal reasoning/net_change/images/RSF-S00008729_t1.png | Temporal reasoning/net_change/images/RSF-S00008729_t2.png | Temporal reasoning/net_change/scene_graphs/RSF-S00008729.json | Temporal reasoning/net_change/net_change.json | [{"claim_type":"Existence","time":"t1","subject":"low_vegetation","value":true,"refs":{"node_ids":["n_t1_low_vegetation"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"low_vegetation","name":"aggregate_area","value":298,"refs":{"node_ids":["n_t1_low_vegetation"],"edge_ids":[]}},{"claim_type":"Existen... | {"program_id":"temporal.max_area_increase","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"low_vegetation"},"outputs":{"exists:low_vegetation:t1":"value"},"refs":{"node_ids":["n_t1_low_vegetation"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"A... |
RSF-Q00000006 | RSF-S00000006 | Perception | object_presence | Does the image contain any helipad? | No | text | [
"Can't judge",
"Yes",
"No"
] | Perception/object_presence/images/RSF-S00000006_t1.jpg | null | Perception/object_presence/scene_graphs/RSF-S00000006.json | Perception/object_presence/object_presence.json | [{"claim_type":"Existence","time":"t1","subject":"helipad","value":false,"refs":{"node_ids":[],"edge_ids":[]}}] | {"program_id":"perception.object_presence","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"helipad"},"outputs":{"answer":"value"},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true}],"answer_fn":{"op":"bool_label","claim_type":"Existence","field":"value","var":"... |
RSF-Q00000011 | RSF-S00000011 | Perception | object_counting | Count the aircraft shown in the image. | 9 | text | [
"8",
"9",
"11",
"10"
] | Perception/object_counting/images/RSF-S00000011_t1.jpg | null | Perception/object_counting/scene_graphs/RSF-S00000011.json | Perception/object_counting/object_counting.json | [{"claim_type":"Existence","time":"t1","subject":"aircraft","value":true,"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003","n_obj_00004","n_obj_00005","n_obj_00006","n_obj_00007","n_obj_00008","n_obj_00009"],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"aircraft","value":9,"refs":{"node_id... | {"program_id":"perception.object_count","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"aircraft","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003","n_obj_00004","n_obj_00005","n_obj_00006","n_obj_00007","n_obj_00008","n_obj_000... |
RSF-Q00000013 | RSF-S00000013 | Perception | object_localization | Is the bridge located in the center region? | Yes | text | [
"Yes",
"No",
"Can't judge"
] | Perception/object_localization/images/RSF-S00000013_t1.jpg | null | Perception/object_localization/scene_graphs/RSF-S00000013.json | Perception/object_localization/object_localization.json | [{"claim_type":"Existence","time":"t1","subject":"bridge","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"bridge","value":"center","refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}}] | {"program_id":"perception.object_location","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"bridge","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Location","scope":{"time":"t1"},"const... |
RSF-Q00001296 | RSF-S00001296 | Perception | fine_grained_recognition | Which subtype best matches the ship? | liquid cargo ship | text | [
"dry cargo ship",
"passenger ship",
"engineering ship",
"liquid cargo ship"
] | Perception/fine_grained_recognition/images/RSF-S00001296_t1.tif | null | Perception/fine_grained_recognition/scene_graphs/RSF-S00001296.json | Perception/fine_grained_recognition/fine_grained_recognition.json | [{"claim_type":"Existence","time":"t1","subject":"ship","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"ship","name":"subtype","value":"liquid_cargo_ship","refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}}] | {"program_id":"perception.object_attribute","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"ship","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"const... |
RSF-Q00004584 | RSF-S00004584 | Relational reasoning | aggregate_distribution | Where are the cargo trucks mainly located? | mostly along the right side of the image | text | [
"mostly along the right side of the image",
"mostly in the lower part of the image",
"mostly near the central area of the image",
"mostly along the left side of the image"
] | Relational reasoning/aggregate_distribution/images/RSF-S00004584_t1.tif | null | Relational reasoning/aggregate_distribution/scene_graphs/RSF-S00004584.json | Relational reasoning/aggregate_distribution/aggregate_distribution.json | [{"claim_type":"Counting","time":"t1","subject":"cargo_truck::top-left","value":1,"refs":{"node_ids":["n_obj_00032"],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"cargo_truck::top-center","value":1,"refs":{"node_ids":["n_obj_00022"],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"cargo... | {"program_id":"spatial.aggregate_distribution","slots":[{"slot_id":"s_001","claim_type":"Counting","scope":{"time":"t1"},"constraints":{"subject":"cargo_truck::top-left","quantity":1},"outputs":{},"refs":{"node_ids":["n_obj_00032"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Counting","scope":{"... |
RSF-Q00004594 | RSF-S00004594 | Relational reasoning | directional | Which listed object is above the vehicle? | the baseball field in the center-left | text | [
"the baseball field in the center",
"the baseball field in the bottom-left",
"the baseball field in the center-left",
"the baseball field in the bottom-right"
] | Relational reasoning/directional/images/RSF-S00004594_t1.jpg | null | Relational reasoning/directional/scene_graphs/RSF-S00004594.json | Relational reasoning/directional/directional.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"baseball_field","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"baseball_field","val... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t1... |
RSF-Q00004687 | RSF-S00004674 | Relational reasoning | proximity | Relative to the vehicle, which option is farthest away? | the baseball field in the center-left | text | [
"the baseball field in the top-right",
"the ground track field",
"the baseball field in the center-left",
"the baseball field in the center-right"
] | Relational reasoning/proximity/images/RSF-S00004674_t1.jpg | null | Relational reasoning/proximity/scene_graphs/RSF-S00004674.json | Relational reasoning/proximity/proximity.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"vehicle","name":"center","value":[659.5,7.5],"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"bas... | {"program_id":"spatial.proximity","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"constraints"... |
RSF-Q00004673 | RSF-S00000516 | Relational reasoning | topological | Which listed object is contained within the outline of the overpass? | the vehicle in the center | text | [
"the vehicle in the center",
"the vehicle in the top-left",
"the vehicle in the bottom-center",
"the vehicle in the top-center"
] | Relational reasoning/topological/images/RSF-S00000516_t1.jpg | null | Relational reasoning/topological/scene_graphs/RSF-S00000516.json | Relational reasoning/topological/topological.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center",... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t... |
RSF-Q00004804 | RSF-S00011447 | Relational reasoning | projective_ordering | What is the correct left-to-right order for the listed aircraft? | the A350, the A220, the A321 | text | [
"the A321, the A220, the A350",
"the A220, the A321, the A350",
"the A350, the A321, the A220",
"the A350, the A220, the A321"
] | Relational reasoning/projective_ordering/images/RSF-S00011447_t1.tif | null | Relational reasoning/projective_ordering/scene_graphs/RSF-S00011447.json | Relational reasoning/projective_ordering/projective_ordering.json | [{"claim_type":"Existence","time":"t1","subject":"a350","value":true,"refs":{"node_ids":["n_obj_00004"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"a350","name":"center","value":[54.0,530.5],"refs":{"node_ids":["n_obj_00004"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"a220","v... | {"program_id":"spatial.projective_ordering","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"a350","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00004"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"const... |
RSF-Q00009481 | RSF-S00008732 | Temporal reasoning | category_turnover | What category appears in the later image but not in the earlier image? | playground | text | [
"playground",
"building",
"non vegetated surface",
"low vegetation"
] | Temporal reasoning/category_turnover/images/RSF-S00008732_t1.png | Temporal reasoning/category_turnover/images/RSF-S00008732_t2.png | Temporal reasoning/category_turnover/scene_graphs/RSF-S00008732.json | Temporal reasoning/category_turnover/category_turnover.json | [{"claim_type":"Existence","time":"t1","subject":"playground","value":false,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","subject":"playground","value":true,"refs":{"node_ids":["n_t2_playgrounds"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t2","subject":"playground","name":"aggrega... | {"program_id":"temporal.category_appearance","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"playground"},"outputs":{"exists:playground:t1":"value"},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"... |
RSF-Q00009470 | RSF-S00008727 | Temporal reasoning | semantic_transition | Which later category receives most of the changed area from earlier low vegetation? | tree | text | [
"non vegetated surface",
"water",
"building",
"tree"
] | Temporal reasoning/semantic_transition/images/RSF-S00008727_t1.png | Temporal reasoning/semantic_transition/images/RSF-S00008727_t2.png | Temporal reasoning/semantic_transition/scene_graphs/RSF-S00008727.json | Temporal reasoning/semantic_transition/semantic_transition.json | [{"claim_type":"Existence","time":"t1","subject":"low_vegetation","value":true,"refs":{"node_ids":["n_t1_low_vegetation"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"low_vegetation","name":"aggregate_area","value":18032,"refs":{"node_ids":["n_t1_low_vegetation"],"edge_ids":[]}},{"claim_type":"Exist... | {"program_id":"temporal.semantic_transition_target","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"low_vegetation","value":true},"outputs":{},"refs":{"node_ids":["n_t1_low_vegetation"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","s... |
RSF-Q00009487 | RSF-S00008735 | Temporal reasoning | net_change | From the earlier image to the later image, which category expands the most in area? | water | text | [
"building",
"non vegetated surface",
"water",
"tree"
] | Temporal reasoning/net_change/images/RSF-S00008735_t1.png | Temporal reasoning/net_change/images/RSF-S00008735_t2.png | Temporal reasoning/net_change/scene_graphs/RSF-S00008735.json | Temporal reasoning/net_change/net_change.json | [{"claim_type":"Existence","time":"t1","subject":"building","value":true,"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"building","name":"aggregate_area","value":26963,"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","sub... | {"program_id":"temporal.max_area_increase","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"building"},"outputs":{"exists:building:t1":"value"},"refs":{"node_ids":["n_t1_buildings"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope"... |
RSF-Q00000012 | RSF-S00000012 | Perception | object_presence | Does the image contain any bridge? | Yes | text | [
"No",
"Can't judge",
"Yes"
] | Perception/object_presence/images/RSF-S00000012_t1.jpg | null | Perception/object_presence/scene_graphs/RSF-S00000012.json | Perception/object_presence/object_presence.json | [{"claim_type":"Existence","time":"t1","subject":"bridge","value":true,"refs":{"node_ids":["n_obj_00002"],"edge_ids":[]}}] | {"program_id":"perception.object_presence","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"bridge"},"outputs":{"answer":"value"},"refs":{"node_ids":["n_obj_00002"],"edge_ids":[]},"sr_required":true}],"answer_fn":{"op":"bool_label","claim_type":"Existence","field":"va... |
RSF-Q00000017 | RSF-S00000017 | Perception | object_counting | Count the harbors shown in the image. | 5 | text | [
"7",
"5",
"4",
"6"
] | Perception/object_counting/images/RSF-S00000017_t1.jpg | null | Perception/object_counting/scene_graphs/RSF-S00000017.json | Perception/object_counting/object_counting.json | [{"claim_type":"Existence","time":"t1","subject":"harbor","value":true,"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003","n_obj_00004","n_obj_00005"],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"harbor","value":5,"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003","n_obj_00004",... | {"program_id":"perception.object_count","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"harbor","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001","n_obj_00002","n_obj_00003","n_obj_00004","n_obj_00005"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002"... |
RSF-Q00000019 | RSF-S00000019 | Perception | object_localization | Which region of the image contains the ship? | center-left | text | [
"top-center",
"top-left",
"center-left",
"center"
] | Perception/object_localization/images/RSF-S00000019_t1.jpg | null | Perception/object_localization/scene_graphs/RSF-S00000019.json | Perception/object_localization/object_localization.json | [{"claim_type":"Existence","time":"t1","subject":"ship","value":true,"refs":{"node_ids":["n_obj_00010"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"ship","value":"center-left","refs":{"node_ids":["n_obj_00010"],"edge_ids":[]}}] | {"program_id":"perception.object_location","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"ship","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00010"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Location","scope":{"time":"t1"},"constra... |
RSF-Q00001298 | RSF-S00001298 | Perception | fine_grained_recognition | Which subtype best matches the large vehicle in the top-center? | cargo truck | text | [
"truck tractor",
"trailer",
"cargo truck",
"dump truck"
] | Perception/fine_grained_recognition/images/RSF-S00001298_t1.tif | null | Perception/fine_grained_recognition/scene_graphs/RSF-S00001298.json | Perception/fine_grained_recognition/fine_grained_recognition.json | [{"claim_type":"Existence","time":"t1","subject":"large_vehicle","value":true,"refs":{"node_ids":["n_obj_00057"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"large_vehicle","value":"top-center","refs":{"node_ids":["n_obj_00057"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"large_v... | {"program_id":"perception.object_attribute","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"large_vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00057"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Location","scope":{"time":"t1"... |
RSF-Q00004585 | RSF-S00004585 | Relational reasoning | aggregate_distribution | Which description best captures where the greenbelts are most concentrated? | mainly in the lower-right part of the image | text | [
"mainly in the lower central part of the image",
"mainly in the upper-right part of the image",
"mainly in the lower-right part of the image",
"mainly near the center of the image"
] | Relational reasoning/aggregate_distribution/images/RSF-S00004585_t1.png | null | Relational reasoning/aggregate_distribution/scene_graphs/RSF-S00004585.json | Relational reasoning/aggregate_distribution/aggregate_distribution.json | [{"claim_type":"Counting","time":"t1","subject":"greenbelt::top-left","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"greenbelt::top-center","value":0,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Counting","time":"t1","subject":"greenbelt::top-right","value":0,"re... | {"program_id":"spatial.aggregate_distribution","slots":[{"slot_id":"s_001","claim_type":"Counting","scope":{"time":"t1"},"constraints":{"subject":"greenbelt::top-left","quantity":0},"outputs":{},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Counting","scope":{"time":"t1"},"co... |
RSF-Q00004595 | RSF-S00004595 | Relational reasoning | directional | Which listed object is below the vehicle? | the storage tank in the bottom-right | text | [
"the storage tank in the center-right",
"the storage tank in the center",
"the storage tank in the top-center",
"the storage tank in the bottom-right"
] | Relational reasoning/directional/images/RSF-S00004595_t1.jpg | null | Relational reasoning/directional/scene_graphs/RSF-S00004595.json | Relational reasoning/directional/directional.json | [{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00006"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"storage_tank","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"storage_tank","value":... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"vehicle","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00006"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t1... |
RSF-Q00004691 | RSF-S00004678 | Relational reasoning | proximity | Which option is farthest from the overpass? | the vehicle in the center | text | [
"the service area in the bottom-center",
"the service area in the center",
"the vehicle in the center",
"the vehicle in the bottom-right"
] | Relational reasoning/proximity/images/RSF-S00004678_t1.jpg | null | Relational reasoning/proximity/scene_graphs/RSF-S00004678.json | Relational reasoning/proximity/proximity.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"overpass","name":"center","value":[783.0,210.0],"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":... | {"program_id":"spatial.proximity","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00003"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"constraints... |
RSF-Q00004684 | RSF-S00004671 | Relational reasoning | topological | Which listed object lies within the outline of the overpass? | the vehicle in the center-right | text | [
"the vehicle in the bottom-left",
"the vehicle in the center",
"the vehicle in the top-center",
"the vehicle in the center-right"
] | Relational reasoning/topological/images/RSF-S00004671_t1.jpg | null | Relational reasoning/topological/scene_graphs/RSF-S00004671.json | Relational reasoning/topological/topological.json | [{"claim_type":"Existence","time":"t1","subject":"overpass","value":true,"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"vehicle","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Location","time":"t1","subject":"vehicle","value":"center-r... | {"program_id":"relational.subject_identification","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"overpass","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00005"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existence","scope":{"time":"t... |
RSF-Q00004828 | RSF-S00004797 | Relational reasoning | projective_ordering | Choose the left-to-right order of the listed aircraft. | the A321, the Boeing 737, the A220 | text | [
"the A220, the A321, the Boeing 737",
"the A220, the Boeing 737, the A321",
"the A321, the A220, the Boeing 737",
"the A321, the Boeing 737, the A220"
] | Relational reasoning/projective_ordering/images/RSF-S00004797_t1.tif | null | Relational reasoning/projective_ordering/scene_graphs/RSF-S00004797.json | Relational reasoning/projective_ordering/projective_ordering.json | [{"claim_type":"Existence","time":"t1","subject":"a321","value":true,"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"a321","name":"center","value":[138.0,99.4999],"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]}},{"claim_type":"Existence","time":"t1","subject":"boein... | {"program_id":"spatial.projective_ordering","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"a321","value":true},"outputs":{},"refs":{"node_ids":["n_obj_00001"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{"time":"t1"},"const... |
RSF-Q00009484 | RSF-S00008733 | Temporal reasoning | category_turnover | Which category can be seen in the later image but not in the earlier image? | non vegetated surface | text | [
"low vegetation",
"tree",
"building",
"non vegetated surface"
] | Temporal reasoning/category_turnover/images/RSF-S00008733_t1.png | Temporal reasoning/category_turnover/images/RSF-S00008733_t2.png | Temporal reasoning/category_turnover/scene_graphs/RSF-S00008733.json | Temporal reasoning/category_turnover/category_turnover.json | [{"claim_type":"Existence","time":"t1","subject":"non_vegetated_surface","value":false,"refs":{"node_ids":[],"edge_ids":[]}},{"claim_type":"Existence","time":"t2","subject":"non_vegetated_surface","value":true,"refs":{"node_ids":["n_t2_NVG_surface"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t2","subject":"non_v... | {"program_id":"temporal.category_appearance","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"non_vegetated_surface"},"outputs":{"exists:non_vegetated_surface:t1":"value"},"refs":{"node_ids":[],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Existe... |
RSF-Q00009473 | RSF-S00008728 | Temporal reasoning | semantic_transition | Which earlier category mainly changes into non vegetated surface in the later image? | playground | text | [
"low vegetation",
"tree",
"water",
"playground"
] | Temporal reasoning/semantic_transition/images/RSF-S00008728_t1.png | Temporal reasoning/semantic_transition/images/RSF-S00008728_t2.png | Temporal reasoning/semantic_transition/scene_graphs/RSF-S00008728.json | Temporal reasoning/semantic_transition/semantic_transition.json | [{"claim_type":"Existence","time":"t1","subject":"playground","value":true,"refs":{"node_ids":["n_t1_playgrounds"],"edge_ids":[]}},{"claim_type":"Attribute","time":"t1","subject":"playground","name":"aggregate_area","value":21261,"refs":{"node_ids":["n_t1_playgrounds"],"edge_ids":[]}},{"claim_type":"Existence","time":"... | {"program_id":"temporal.semantic_transition_source","slots":[{"slot_id":"s_001","claim_type":"Existence","scope":{"time":"t1"},"constraints":{"subject":"playground","value":true},"outputs":{},"refs":{"node_ids":["n_t1_playgrounds"],"edge_ids":[]},"sr_required":true},{"slot_id":"s_002","claim_type":"Attribute","scope":{... |
RSFaith-Bench
RSFaith-Bench is a remote-sensing vision-language benchmark designed to evaluate grounded visual reasoning beyond surface-level object recognition. The benchmark covers perception, relational reasoning, and temporal reasoning over remote-sensing imagery. Each example is formulated as a multiple-choice question and is paired with a compact scene graph, supporting evidence, and an executable reasoning program.
The release contains 13,511 question-answer records, 16,288 referenced images, and 12,876 compact scene graphs.
Using the Dataset
The annotation files can be loaded directly as JSON. The root-level
metadata.jsonl provides a flat index over all records:
from datasets import load_dataset
dataset = load_dataset(
"json",
data_files={"benchmark": "metadata.jsonl"},
split="benchmark",
)
print(dataset[0])
Images and scene graphs are stored in per-subcategory archives. To restore the file layout referenced by the JSON records, extract the archives in place:
huggingface-cli download <namespace>/RSFaith-Bench \
--repo-type dataset \
--local-dir RSFaith-Bench
find RSFaith-Bench -name assets.tar.zst -print0 |
while IFS= read -r -d '' archive; do
(cd "$(dirname "$archive")" && tar -I zstd -xf assets.tar.zst)
done
After extraction, the images and scene_graph fields in each subcategory
JSON file resolve relative to that subcategory directory. The image_t1,
image_t2, and scene_graph fields in metadata.jsonl resolve relative to
the repository root.
File Organization
The dataset is organized by reasoning level and subcategory. Each subcategory directory contains:
<subcategory>.json: question-answer records for the subcategory.assets.tar.zst: compressedimages/andscene_graphs/directories.
RSFaith-Bench
βββ README.md
βββ metadata.jsonl
βββ dataset_manifest.json
βββ croissant.json
βββ Perception
β βββ object_presence
β β βββ object_presence.json
β β βββ assets.tar.zst
β βββ object_counting
β β βββ object_counting.json
β β βββ assets.tar.zst
β βββ fine_grained_recognition
β β βββ fine_grained_recognition.json
β β βββ assets.tar.zst
β βββ object_localization
β βββ object_localization.json
β βββ assets.tar.zst
βββ Relational reasoning
β βββ directional
β β βββ directional.json
β β βββ assets.tar.zst
β βββ topological
β β βββ topological.json
β β βββ assets.tar.zst
β βββ proximity
β β βββ proximity.json
β β βββ assets.tar.zst
β βββ projective_ordering
β β βββ projective_ordering.json
β β βββ assets.tar.zst
β βββ aggregate_distribution
β βββ aggregate_distribution.json
β βββ assets.tar.zst
βββ Temporal reasoning
βββ category_turnover
β βββ category_turnover.json
β βββ assets.tar.zst
βββ net_change
β βββ net_change.json
β βββ assets.tar.zst
βββ semantic_transition
βββ semantic_transition.json
βββ assets.tar.zst
Data Fields
Each question-answer record contains the following fields:
question_id: anonymized question identifier.scene_id: anonymized scene identifier.level: high-level reasoning category.subcategory: fine-grained reasoning category.question: natural-language question.answer: correct answer.answer_type: answer representation.choices: multiple-choice options.images: relative image paths.scene_graph: relative scene graph path.support: grounded support evidence.program: executable reasoning specification.
Dataset Statistics
| Level | Subcategory | Records |
|---|---|---|
| Perception | object_presence | 969 |
| Perception | object_counting | 1,085 |
| Perception | fine_grained_recognition | 1,228 |
| Perception | object_localization | 1,298 |
| Relational reasoning | directional | 1,166 |
| Relational reasoning | topological | 921 |
| Relational reasoning | proximity | 987 |
| Relational reasoning | projective_ordering | 944 |
| Relational reasoning | aggregate_distribution | 866 |
| Temporal reasoning | category_turnover | 1,515 |
| Temporal reasoning | net_change | 1,273 |
| Temporal reasoning | semantic_transition | 1,259 |
Dataset Construction
RSFaith-Bench is constructed from remote-sensing scenes represented as grounded scene graphs. The scene graphs encode objects, spatial relations, temporal changes, and compact global inventories when applicable. Question-answer pairs are generated from programmatic templates and then curated to balance reasoning categories, answer distributions, and scene coverage. The released records retain the reasoning support and program specification so that each answer can be traced back to the corresponding scene graph.
Licensing
The dataset is released under the Creative Commons Attribution Non Commercial 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Citation Information
If you use RSFaith-Bench in your research, please cite the accompanying paper:
@misc{rsfaithbench2026,
title = {RSFaith-Bench: When Correct Answers Come with Unfaithful Evidence in Remote Sensing MLLMs},
author = {Anonymous},
year = {2026}
}
Acknowledgement
RSFaith-Bench is built from remote-sensing data sources including DIOR, DOTA, FAIR1M, SECOND, xBD, and ReCon1M. We thank the creators and maintainers of these datasets for making their resources available to the research community.
- Downloads last month
- 17