Triple
T9734918
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Carnuntum |
E236032
|
entity |
| Predicate | hasSite |
P1205
|
FINISHED |
| Object |
Heidentor
Heidentor is a monumental late Roman triumphal arch located near the ancient city of Carnuntum in present-day Austria.
|
E818613
|
NE FINISHED |
How this triple was built (4 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: Heidentor | Statement: [Carnuntum, hasSite, Heidentor]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Heidentor Context triple: [Carnuntum, hasSite, Heidentor]
-
A.
Stadttor
Stadttor is a prominent modern office and government building in Düsseldorf, Germany, known for its distinctive glass architecture and role as a landmark of the MedienHafen area.
-
B.
Torhalle
Torhalle is an early medieval gatehouse at Lorsch Abbey in Germany, renowned as one of the best-preserved examples of Carolingian architecture.
-
C.
Burgtor
Burgtor is a historic city gate in Lübeck, Germany, notable as part of the city’s medieval fortifications and Hanseatic architectural heritage.
-
D.
Ellinger Tor
Ellinger Tor is a historic city gate and one of the best-known architectural landmarks of the medieval town of Weißenburg in Bayern, Germany.
-
E.
Valkhof
Valkhof is a historic site in Nijmegen, Netherlands, known for its hilltop park and medieval castle ruins overlooking the River Waal.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg
Description generation
gpt-5.1
Instruction
Generate a one-sentence description of the target entity. You are given a context triple in the form (subject, predicate, object), where the object is the target entity. # Instructions Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. Avoid repeating the information from the triple, unless really essential. # Response Format Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: Heidentor Triple: [Carnuntum, hasSite, Heidentor]
Generated description
Heidentor is a monumental late Roman triumphal arch located near the ancient city of Carnuntum in present-day Austria.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: Heidentor Target entity description: Heidentor is a monumental late Roman triumphal arch located near the ancient city of Carnuntum in present-day Austria.
-
A.
Stadttor
Stadttor is a prominent modern office and government building in Düsseldorf, Germany, known for its distinctive glass architecture and role as a landmark of the MedienHafen area.
-
B.
Torhalle
Torhalle is an early medieval gatehouse at Lorsch Abbey in Germany, renowned as one of the best-preserved examples of Carolingian architecture.
-
C.
Burgtor
Burgtor is a historic city gate in Lübeck, Germany, notable as part of the city’s medieval fortifications and Hanseatic architectural heritage.
-
D.
Ellinger Tor
Ellinger Tor is a historic city gate and one of the best-known architectural landmarks of the medieval town of Weißenburg in Bayern, Germany.
-
E.
Valkhof
Valkhof is a historic site in Nijmegen, Netherlands, known for its hilltop park and medieval castle ruins overlooking the River Waal.
- F. None of above. chosen
Provenance (5 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_69ca84d313e88190983ee6ffd0ef60d2 |
completed | March 30, 2026, 2:12 p.m. |
| NER | Named-entity recognition | batch_69cd9eebcec08190a9d4606fd26f2e19 |
completed | April 1, 2026, 10:40 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69d1afce9834819090949690b4ca8622 |
completed | April 5, 2026, 12:41 a.m. |
| NEDg | Description generation | batch_69d1b098c3348190a298cfe4921c8921 |
completed | April 5, 2026, 12:45 a.m. |
| NED2 | Entity disambiguation (via description) | batch_69d1b124659481909e7a2ecaf01d8a50 |
completed | April 5, 2026, 12:47 a.m. |
Created at: March 30, 2026, 8:22 p.m.