Triple
T35482604
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Lord of Harrenhal |
E1025505
|
entity |
| Predicate | associatedWithStructureTrait |
P195670
|
FINISHED |
| Object | vast castle |
—
|
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: vast castle | Statement: [Lord of Harrenhal, associatedWithStructureTrait, vast castle]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: associatedWithStructureTrait Context triple: [Lord of Harrenhal, associatedWithStructureTrait, vast castle]
-
A.
associatesStructure
Indicates a relationship where one entity is linked or connected to a particular structure, typically to denote that the structure is relevant, attached, or functionally associated with it.
-
B.
associatedWithDefiningTrait
Indicates that one entity is linked to another as a key characteristic or defining feature of it.
-
C.
associatedWithAttribute
Indicates that one entity is linked to, characterized by, or described through a particular attribute.
-
D.
associatedWithVehicleOrStructure
chosen
Indicates that an entity has a relevant connection or involvement with a particular vehicle or structure.
-
E.
associatedWithPersonaTrait
Indicates that an entity is linked to or characterized by a particular persona trait or attribute.
- 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_69f76dfbcdd881908c7b0b6bc502252b |
completed | May 3, 2026, 3:47 p.m. |
| NER | Named-entity recognition | batch_6a037c8d06cc8190ab6a5e18d9d2571e |
completed | May 12, 2026, 7:16 p.m. |
| PD | Predicate disambiguation | batch_6a037a04d8348190a4819666eab42c9b |
completed | May 12, 2026, 7:05 p.m. |
Created at: May 3, 2026, 4:04 p.m.