Triple
T3688532
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Leipzig-Thekla subcamp |
E78283
|
entity |
| Predicate | heldPrisonerGroup |
P41072
|
FINISHED |
| Object | political prisoners |
—
|
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: political prisoners | Statement: [Leipzig-Thekla subcamp, heldPrisonerGroup, political prisoners]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: heldPrisonerGroup Context triple: [Leipzig-Thekla subcamp, heldPrisonerGroup, political prisoners]
-
A.
imprisonedWith
Indicates that two entities are confined or held in prison together at the same time and place.
-
B.
estimatedPrisonerCount
Indicates the estimated number of prisoners associated with a particular context, such as a location, time period, or event.
-
C.
hasPrison
Indicates that one entity possesses, contains, or is the location of a prison associated with another entity.
-
D.
accusedGroup
Indicates that one group is formally charged or blamed by another party for committing a wrongdoing or offense.
-
E.
inmates
chosen
Indicates that one entity is confined or held as a prisoner within an institution or facility associated with another entity.
- 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_69ad85e285a081908f8cbfa9e2ed9b75 |
completed | March 8, 2026, 2:21 p.m. |
| NER | Named-entity recognition | batch_69adc4c960788190b73ede08658846aa |
completed | March 8, 2026, 6:49 p.m. |
| PD | Predicate disambiguation | batch_69adb84be1fc81909721c871babb4633 |
completed | March 8, 2026, 5:56 p.m. |
Created at: March 8, 2026, 3:26 p.m.