Triple
T36958675
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Yellow Armor |
E914251
|
entity |
| Predicate | canBeDeniedBy |
P145804
|
FINISHED |
| Object | opponents picking it up first |
—
|
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: opponents picking it up first | Statement: [Yellow Armor, canBeDeniedBy, opponents picking it up first]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: canBeDeniedBy Context triple: [Yellow Armor, canBeDeniedBy, opponents picking it up first]
-
A.
isDeniedBy
chosen
Indicates that a request, claim, or action directed at an entity is explicitly refused or rejected by another entity.
-
B.
deniedTo
Indicates that access, permission, or a requested benefit is explicitly refused or withheld from a particular entity.
-
C.
canBeDeniedForAdverseCredit
Indicates that something (such as an application or request) may be refused on the basis of negative or unfavorable credit information.
-
D.
canRejectWith
Indicates that one entity has the authority or ability to refuse, decline, or dismiss another entity or proposal using a specific method or condition.
-
E.
deniedAs
Indicates that one entity has refused, rejected, or not granted a request, claim, or association involving 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_69f76e8c498c8190b2842db80aea8b3b |
completed | May 3, 2026, 3:49 p.m. |
| NER | Named-entity recognition | batch_69feff70fbec8190b1ff5f943f29613e |
completed | May 9, 2026, 9:33 a.m. |
| PD | Predicate disambiguation | batch_69fefbcd5b7881909cfe52b32f8a4301 |
completed | May 9, 2026, 9:18 a.m. |
Created at: May 3, 2026, 4:13 p.m.