Triple

T23304787
Position Surface form Disambiguated ID Type / Status
Subject Battle of Zahle E590404 entity
Predicate locatedIn P40 FINISHED
Object Zahle NE NERFINISHED

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: Zahle | Statement: [Battle of Zahle, locatedIn, Zahle]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Zahle
Context triple: [Battle of Zahle, locatedIn, Zahle]
  • A. Zahle chosen
    Zahle is a prominent city in Lebanon’s Beqaa Valley, known for its historic architecture, vineyards, and role as a regional commercial and cultural center.
  • B. Havelberg
    Havelberg is a small historic town in Saxony-Anhalt, Germany, known for its medieval cathedral and location at the confluence of the Havel and Elbe rivers.
  • C. Hardegsen
    Hardegsen is a small town in Lower Saxony, Germany, known for its medieval castle and historic town center.
  • D. Grenaa
    Grenaa is a coastal town in eastern Jutland, Denmark, known for its ferry connections to the island of Anholt and its role as a regional commercial and educational center.
  • E. Haderup
    Haderup is a small town in Denmark, known locally as a rural community that gave its name to the former Aulum-Haderup Municipality.
  • F. None of above.
  • G. Unsure - the case is ambiguous/there is not enough information to decide.

Provenance (2 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_69e25d1c0ecc8190a355aa229f06d0e0 completed April 17, 2026, 4:17 p.m.
NER Named-entity recognition batch_69f19725b248819089e61efb82e3440f completed April 29, 2026, 5:29 a.m.
Created at: April 17, 2026, 5:04 p.m.