Triple
T1569430
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Bengal Sultanate |
E33505
|
entity |
| Predicate | capital |
P234
|
FINISHED |
| Object | Gaur |
E167994
|
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: Gaur | Statement: [Bengal Sultanate, capital, Gaur]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Gaur Context triple: [Bengal Sultanate, capital, Gaur]
-
A.
Gaur
chosen
Gaur is a ruined medieval city on the India–Bangladesh border, historically a major capital of Bengal known for its Islamic architecture and archaeological remains.
-
B.
Nilgiri tahr
The Nilgiri tahr is an endangered wild mountain goat endemic to the high-altitude grasslands of southern India, known for its curved horns and rugged cliff-dwelling lifestyle.
-
C.
Derawali
Derawali is a regional dialect of the Saraiki language spoken primarily in and around the Dera Ghazi Khan and Dera Ismail Khan areas of Pakistan.
-
D.
Indian rhinoceros
The Indian rhinoceros is a large, single-horned rhinoceros native to the Indian subcontinent, known for its thick, armor-like skin and conservation-dependent populations in protected areas.
-
E.
Lion
Lion is a 2016 biographical drama film about an Indian boy separated from his family and adopted in Australia who later uses Google Earth to find his way home.
- 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_69a885f11b048190935025a035302715 |
completed | March 4, 2026, 7:20 p.m. |
| NER | Named-entity recognition | batch_69a908b67304819081ad555000e51197 |
completed | March 5, 2026, 4:38 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69ad4023da008190a947cf0d2983df2e |
completed | March 8, 2026, 9:23 a.m. |
Created at: March 4, 2026, 7:27 p.m.