Triple
T6798919
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Russian Second Army |
E156130
|
entity |
| Predicate | fateOfCommanderAtTannenberg |
P73058
|
FINISHED |
| Object | Alexander Samsonov committed suicide |
—
|
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: Alexander Samsonov committed suicide | Statement: [Russian Second Army, fateOfCommanderAtTannenberg, Alexander Samsonov committed suicide]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: fateOfCommanderAtTannenberg Context triple: [Russian Second Army, fateOfCommanderAtTannenberg, Alexander Samsonov committed suicide]
-
A.
PrussianCommander
Indicates that an individual holds the position or role of a commander within the Prussian military forces.
-
B.
notableCommanderKilledInAction
Indicates that a notable military commander associated with an entity was killed in action.
-
C.
alliedCommanderKilled
Indicates that a commander belonging to an allied force has been killed.
-
D.
notableCommanderEnd
Indicates that a person served as a notable commander of an entity (such as a military unit or organization) at the end of a specified period or timeframe.
-
E.
germanCommander
Indicates that the subject serves as a military commander for German forces in relation to the object.
- F. None of above. chosen
Provenance (4 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_69c6881844448190a65822d9b39d7f88 |
completed | March 27, 2026, 1:37 p.m. |
| NER | Named-entity recognition | batch_69c6d2cb6b2881909b30bb8020a9d3bf |
completed | March 27, 2026, 6:56 p.m. |
| PD | Predicate disambiguation | batch_69c6d099bf08819089a9f9894d037e74 |
completed | March 27, 2026, 6:46 p.m. |
| PDg | Predicate description generation | batch_69c6d2a8f9188190abbb8c730e7b5edf |
completed | March 27, 2026, 6:55 p.m. |
Created at: March 27, 2026, 2:15 p.m.