Triple
T14416022
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Davos Seaworth |
E357451
|
entity |
| Predicate | keepsAsToken |
P35371
|
FINISHED |
| Object | finger bones in a pouch |
—
|
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: finger bones in a pouch | Statement: [Davos Seaworth, keepsAsToken, finger bones in a pouch]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: keepsAsToken Context triple: [Davos Seaworth, keepsAsToken, finger bones in a pouch]
-
A.
isKeptAs
Indicates that one entity is maintained, preserved, or retained by another entity, often implying possession, storage, or ongoing custody.
-
B.
keptIn
chosen
Indicates that one entity is stored, contained, or maintained inside another entity as its location or container.
-
C.
keptWith
Indicates that one entity is stored, maintained, or held together in association with another entity.
-
D.
isKeptFor
Indicates that one entity is retained or preserved for the benefit, use, or purpose of another entity.
-
E.
nativeToken
Indicates that the referenced token is the primary, built-in cryptocurrency or asset of a given blockchain or platform, as opposed to a secondary or issued token.
- 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_69d82793421c8190861eb0e673b085de |
completed | April 9, 2026, 10:26 p.m. |
| NER | Named-entity recognition | batch_69de90cc99208190a2313b1acfb5d802 |
completed | April 14, 2026, 7:09 p.m. |
| PD | Predicate disambiguation | batch_69de5c30467881908e770e3940295641 |
completed | April 14, 2026, 3:24 p.m. |
Created at: April 10, 2026, 1:17 a.m.