Triple

T3842280
Position Surface form Disambiguated ID Type / Status
Subject Meuse E93477 entity
Predicate contains P35 FINISHED
Object Fort Vaux E100025 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: Fort Vaux | Statement: [Meuse, contains, Fort Vaux]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Fort Vaux
Context triple: [Meuse, contains, Fort Vaux]
  • A. Fort Vaux chosen
    Fort Vaux is a French World War I fortress near Verdun, renowned for its fierce 1916 defense against German forces during the Battle of Verdun.
  • B. Fort du Roule
    Fort du Roule is a coastal fortification overlooking Cherbourg in Normandy, France, notable for its role in World War II and its commanding views over the English Channel.
  • C. Fort Conde
    Fort Conde is a reconstructed 18th-century French colonial fort and museum in Mobile, Alabama, that interprets the region’s early military and colonial history.
  • D. Bourlon Barracks
    Bourlon Barracks is a British Army installation within Catterick Garrison in North Yorkshire, used for housing and training military units.
  • E. Fort Carré
    Fort Carré is a 16th-century star-shaped coastal fortress in Antibes, France, historically built to defend the town and its harbor and now a prominent historic monument overlooking the Mediterranean.
  • 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_69aed96ce578819084ab16e3439976c9 completed March 9, 2026, 2:30 p.m.
NER Named-entity recognition batch_69aeebb397ac81908f74a42a0eeb8682 completed March 9, 2026, 3:48 p.m.
NED1 Entity disambiguation (via context triple) batch_69b5040d478081909a903bbf02f0d0ec completed March 14, 2026, 6:45 a.m.
Created at: March 9, 2026, 3:18 p.m.