Triple
T3790935
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Third Battle of Kharkov |
E89643
|
entity |
| Predicate | cityRecapturedBy |
P34167
|
FINISHED |
| Object | German forces |
—
|
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: German forces | Statement: [Third Battle of Kharkov, cityRecapturedBy, German forces]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: cityRecapturedBy Context triple: [Third Battle of Kharkov, cityRecapturedBy, German forces]
-
A.
wasRecapturedBy
chosen
Indicates that an entity which had previously escaped or been released was caught again by another entity.
-
B.
cityHeld
Indicates that a particular city was the location where an event, activity, or occurrence took place or was hosted.
-
C.
citySacked
Indicates that a city has been violently attacked, looted, and destroyed or heavily damaged, typically as a result of military conquest.
-
D.
conqueredCity
Indicates that one entity has taken control of a city from another entity, typically through military force or domination.
-
E.
cityOwned
Indicates that something is owned, managed, or under the jurisdiction of a city or municipal government.
- 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_69aed9597d6881909b6ee3b9de859223 |
completed | March 9, 2026, 2:29 p.m. |
| NER | Named-entity recognition | batch_69aeecefa3608190a7a20ed6df6a64b2 |
completed | March 9, 2026, 3:53 p.m. |
| PD | Predicate disambiguation | batch_69aee743c8d08190a9f9c97b836bd703 |
completed | March 9, 2026, 3:29 p.m. |
Created at: March 9, 2026, 3:15 p.m.