Triple
T38005558
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Epanterias |
E948224
|
entity |
| Predicate | likelySynonymOf |
P107996
|
FINISHED |
| Object | Allosaurus |
—
|
NE NERFINISHED |
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: Allosaurus | Statement: [Epanterias, likelySynonymOf, Allosaurus]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: likelySynonymOf Context triple: [Epanterias, likelySynonymOf, Allosaurus]
-
A.
synonym
Indicates that two terms have the same or nearly the same meaning in a given context.
-
B.
likelyVariantOf
Indicates that one entity is probably a variant, version, or alternative form of another entity, based on available evidence or similarity.
-
C.
heteronymOf
Indicates that two words share the same spelling but differ in pronunciation and meaning.
-
D.
lessSimilarTo
Indicates that one entity is considered to share fewer similarities or a weaker resemblance with another entity compared to some reference or alternative.
-
E.
oftenEquatedWith
chosen
Indicates that one entity is frequently regarded, interpreted, or treated as being the same as or equivalent to another entity, though not necessarily strictly identical.
- 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_69f76efb4b10819092c8c2ba28ac06a8 |
completed | May 3, 2026, 3:51 p.m. |
| NER | Named-entity recognition | batch_69fbc995dc2481908b3bd4217f8101e7 |
completed | May 6, 2026, 11:07 p.m. |
| PD | Predicate disambiguation | batch_69fbc8ee04f08190977b7ad70fc85896 |
completed | May 6, 2026, 11:04 p.m. |
Created at: May 3, 2026, 4:20 p.m.