Triple
T2829226
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Battle of Uman |
E62195
|
entity |
| Predicate | natureOfDefeat |
P26452
|
FINISHED |
| Object | large Soviet encirclement |
—
|
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: large Soviet encirclement | Statement: [Battle of Uman, natureOfDefeat, large Soviet encirclement]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: natureOfDefeat Context triple: [Battle of Uman, natureOfDefeat, large Soviet encirclement]
-
A.
defeatedBy
Indicates that one entity has been beaten, overcome, or conquered by another in a contest, conflict, or competition.
-
B.
defeat
Indicates that one entity wins against and overcomes another in a contest, conflict, or competition.
-
C.
defeatedCandidate
Indicates that one candidate has won an election or contest against another candidate, causing the other to lose.
-
D.
causeOfDownfall
chosen
Indicates a factor, event, or agent that brings about the failure, ruin, or collapse of someone or something.
-
E.
loserScore
Indicates the number of points or score achieved by the losing participant in a competitive event or comparison.
- 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_69ab4c3c39188190955b9c49d98463d8 |
completed | March 6, 2026, 9:50 p.m. |
| NER | Named-entity recognition | batch_69abde97168c8190b31122b2ad9fdebf |
completed | March 7, 2026, 8:15 a.m. |
| PD | Predicate disambiguation | batch_69abdd0acab881909e8c25cbef83678c |
completed | March 7, 2026, 8:08 a.m. |
Created at: March 6, 2026, 10:01 p.m.