Triple
T127931
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Gold Beach |
E2589
|
entity |
| Predicate | fortifications |
P1090
|
FINISHED |
| Object | German coastal defenses |
—
|
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: German coastal defenses | Statement: [Gold Beach, fortifications, German coastal defenses]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: fortifications Context triple: [Gold Beach, fortifications, German coastal defenses]
-
A.
castleBuiltFor
Indicates that a castle was constructed specifically for the benefit, use, or commission of a particular person, group, or purpose.
-
B.
historicalStronghold
chosen
Indicates that a location has historically served as a fortified center of power, defense, or control for a group or authority.
-
C.
defender
Indicates a relationship where one entity protects, guards, or supports another entity against threats, attacks, or criticism.
-
D.
airDefenseSystemComponent
Indicates that one entity is a component or subsystem of an air defense system associated with another entity.
-
E.
currentStronghold
Indicates that an entity currently serves as the primary base, fortress, or central stronghold for another entity.
- 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_69a2520c0f3481908b0ed054a2fca8d0 |
completed | Feb. 28, 2026, 2:25 a.m. |
| NER | Named-entity recognition | batch_69a25763ccf8819094e8dffb2ff98480 |
completed | Feb. 28, 2026, 2:48 a.m. |
| PD | Predicate disambiguation | batch_69a2564c11208190ad25495609d94d87 |
completed | Feb. 28, 2026, 2:43 a.m. |
Created at: Feb. 28, 2026, 2:30 a.m.