Triple
T26707581
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | chimney swift |
E673321
|
entity |
| Predicate | roostingPosture |
P161148
|
FINISHED |
| Object | clings to vertical surfaces |
—
|
LITERAL FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: clings to vertical surfaces | Statement: [chimney swift, roostingPosture, clings to vertical surfaces]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: roostingPosture Context triple: [chimney swift, roostingPosture, clings to vertical surfaces]
-
A.
roostType
Indicates the type or kind of roost (e.g., structure or location) an animal uses for resting or nesting.
-
B.
perchingSite
Indicates a location or object where an entity (typically an animal, such as a bird) rests or sits temporarily.
-
C.
restingBehavior
Indicates a relationship where an entity is in a state of rest or inactivity, typically pausing movement or action for recovery or idling.
-
D.
baskingBehavior
Indicates engaging in a behavior of lying or resting in warmth or light, typically to absorb heat or comfort.
-
E.
birdPosition
Indicates the spatial location or placement of a bird relative to a reference frame or environment.
- F. None of above. chosen
Provenance (4 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69eecda3a22881908f3061c760b9d542 |
completed | April 27, 2026, 2:44 a.m. |
| NER | Named-entity recognition | batch_69f617bb5a38819099dace3392694896 |
completed | May 2, 2026, 3:26 p.m. |
| PD | Predicate disambiguation | batch_69f60b8dfa0c8190864e1a940024d0a0 |
completed | May 2, 2026, 2:34 p.m. |
| PDg | Predicate description generation | batch_69f6106d346c8190868489f36c65b6ec |
completed | May 2, 2026, 2:55 p.m. |
Created at: April 27, 2026, 3:34 a.m.