Triple
T11382661
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Battle of the Sea of Azov |
E269634
|
entity |
| Predicate | axisCasualties |
P1399
|
FINISHED |
| Object | relatively light compared to Soviet losses |
—
|
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: relatively light compared to Soviet losses | Statement: [Battle of the Sea of Azov, axisCasualties, relatively light compared to Soviet losses]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: axisCasualties Context triple: [Battle of the Sea of Azov, axisCasualties, relatively light compared to Soviet losses]
-
A.
nativeCasualties
Indicates that native or indigenous people suffered deaths or injuries as a result of a particular event, action, or conflict.
-
B.
casualties
chosen
Indicates that an event, action, or situation resulted in people being killed or injured.
-
C.
primaryCasualtiesFrom
Indicates that an entity is the main source or cause of the casualties experienced by another entity.
-
D.
casualtiesAssociatedWithEvent
Indicates that certain casualties (deaths or injuries) are linked to, or resulted from, a specific event.
-
E.
casualtiesCountry
Indicates that the specified country is the one in which the recorded casualties (deaths or injuries) occurred or to which those casualties belong.
- 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_69d6aacca1048190b39dbbc2174616fa |
completed | April 8, 2026, 7:21 p.m. |
| NER | Named-entity recognition | batch_69d800160a1c81909d115bf89fe54a49 |
completed | April 9, 2026, 7:37 p.m. |
| PD | Predicate disambiguation | batch_69d7e70b228c8190b87f5101fd683788 |
completed | April 9, 2026, 5:51 p.m. |
Created at: April 8, 2026, 9:34 p.m.