Triple
T10127190
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Karaims |
E226244
|
entity |
| Predicate | historicalCenter |
P2536
|
FINISHED |
| Object | Trakai |
E286925
|
NE 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: Trakai | Statement: [Karaims, historicalCenter, Trakai]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Trakai Context triple: [Karaims, historicalCenter, Trakai]
-
A.
Trakai
chosen
Trakai is a historic Lithuanian town famed for its medieval island castle and former status as a political center of the Grand Duchy of Lithuania.
-
B.
Marijampolė
Marijampolė is a city in southern Lithuania that serves as an important regional center for administration, culture, and industry.
-
C.
Vilkaviškis
Vilkaviškis is a town in southwestern Lithuania known as an administrative and historical center of the surrounding agricultural region.
-
D.
Mitau
Mitau, historically known as the capital of the Duchy of Courland and Semigallia, is the former German name for the city now called Jelgava in present-day Latvia.
-
E.
Švenčionys
Švenčionys is a small historic town in eastern Lithuania known for its multicultural past and former Jewish community.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
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_69ca843057b48190a86730167f5d6b98 |
completed | March 30, 2026, 2:09 p.m. |
| NER | Named-entity recognition | batch_69cdd2eef7388190b95ffd02814f2d1f |
completed | April 2, 2026, 2:22 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69d2cc69a5c88190ab7b108e1aab20ba |
completed | April 5, 2026, 8:56 p.m. |
Created at: March 30, 2026, 9:05 p.m.