Triple
T5012337
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Salzburg Cathedral |
E112653
|
entity |
| Predicate | previousStyleOnSite |
P30495
|
FINISHED |
| Object | Romanesque |
—
|
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: Romanesque | Statement: [Salzburg Cathedral, previousStyleOnSite, Romanesque]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: previousStyleOnSite Context triple: [Salzburg Cathedral, previousStyleOnSite, Romanesque]
-
A.
primaryStyle
Indicates the main or predominant style associated with an entity, distinguishing it from other secondary or supporting styles.
-
B.
styleWhenFormer
chosen
Indicates a stylistic attribute or manner that applied to an entity during a previous or former state, role, or period.
-
C.
formerSiteOf
Indicates that a location previously hosted or contained something (such as a structure, organization, or event) that is no longer present there.
-
D.
primaryBodyStyle
Indicates the main or predominant body configuration or structural form associated with an entity.
-
E.
previousMode
Indicates that one mode directly precedes another in a sequence or state transition.
- F. None of above.
Provenance (3 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_69bd4434acb8819086679dbeccc2fe54 |
completed | March 20, 2026, 12:57 p.m. |
| NER | Named-entity recognition | batch_69bd730f12a481908a27c15dc73987c6 |
completed | March 20, 2026, 4:17 p.m. |
| PD | Predicate disambiguation | batch_69bd714cbc448190aa53a8a83d768b64 |
completed | March 20, 2026, 4:09 p.m. |
Created at: March 20, 2026, 1:35 p.m.