Triple
T34639984
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Havens of the Falas |
E889528
|
entity |
| Predicate | refugeFor |
P74118
|
FINISHED |
| Object | survivors of the wars of Beleriand |
—
|
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: survivors of the wars of Beleriand | Statement: [Havens of the Falas, refugeFor, survivors of the wars of Beleriand]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: refugeFor Context triple: [Havens of the Falas, refugeFor, survivors of the wars of Beleriand]
-
A.
takenAsRefugeBy
Indicates that one entity is used or relied upon by another entity as a place or source of safety, protection, or refuge.
-
B.
tookRefugeWith
Indicates that one entity sought safety, protection, or shelter by going to and staying with another entity.
-
C.
designatedAsRefuge
chosen
Indicates that an entity has been officially assigned or recognized as a refuge or place of protection.
-
D.
seeksRefugeFrom
Indicates that one entity actively looks for safety, protection, or shelter in order to escape or avoid another entity.
-
E.
relativeShelterFrom
Indicates that one entity provides shelter or protection for another entity relative to some reference condition, location, or alternative.
- 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_69f349d724848190b63ad3407e0006d9 |
completed | April 30, 2026, 12:23 p.m. |
| NER | Named-entity recognition | batch_69f72291022081909c00d6aec5098b50 |
completed | May 3, 2026, 10:25 a.m. |
| PD | Predicate disambiguation | batch_69f72157af108190880317a62e634bb0 |
completed | May 3, 2026, 10:20 a.m. |
Created at: May 1, 2026, 2:04 a.m.