Triple

T4133143
Position Surface form Disambiguated ID Type / Status
Subject Dunhuang E85086 entity
Predicate locatedIn P40 FINISHED
Object Hexi Corridor E399464 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: Hexi Corridor | Statement: [Dunhuang, locatedIn, Hexi Corridor]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Hexi Corridor
Context triple: [Dunhuang, locatedIn, Hexi Corridor]
  • A. Hexi Corridor chosen
    The Hexi Corridor is a historically vital stretch of the Silk Road in northwestern China, serving as a strategic gateway between central China and Central Asia.
  • B. Hexi District
    Hexi District is a central urban district of Tianjin, China, known for its commercial areas, educational institutions, and cultural sites along the Hai River.
  • C. Changling
    Changling is the largest and best-preserved mausoleum within Beijing’s Ming Tombs complex, built for the Yongle Emperor and his empress.
  • D. Juyongguan
    Juyongguan is a famous mountain pass northwest of Beijing that hosts one of the most visited and historically significant sections of the Great Wall of China.
  • E. Xijing
    Xijing is the historical name of Xi'an, one of China’s oldest and most important ancient capitals.
  • 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_69aed935ccd881909dc61f81bcdb7a78 completed March 9, 2026, 2:29 p.m.
NER Named-entity recognition batch_69af02306d608190bf8945d560f87c21 completed March 9, 2026, 5:24 p.m.
NED1 Entity disambiguation (via context triple) batch_69b576c4c0888190be156d093ea11207 completed March 14, 2026, 2:55 p.m.
Created at: March 9, 2026, 3:43 p.m.