Triple

T14732978
Position Surface form Disambiguated ID Type / Status
Subject Manchester Avenue E346124 entity
Predicate hasJunctionWith P1018 FINISHED
Object Vermont Avenue E119130 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: Vermont Avenue | Statement: [Manchester Avenue, hasJunctionWith, Vermont Avenue]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Vermont Avenue
Context triple: [Manchester Avenue, hasJunctionWith, Vermont Avenue]
  • A. Vermont Avenue chosen
    Vermont Avenue is a major north–south thoroughfare in Los Angeles, California, running through several neighborhoods and serving as a key urban artery.
  • B. Wyoming Avenue
    Wyoming Avenue is a significant thoroughfare in Philadelphia, Pennsylvania, serving as a key corridor for local traffic and public transportation.
  • C. Vermont Avenue NW
    Vermont Avenue NW is a major diagonal street in Washington, D.C., running northwest from downtown through neighborhoods such as Logan Circle.
  • D. Constitution Avenue
    Constitution Avenue is a major thoroughfare in Abuja, Nigeria, running through the city’s central government district and serving key national institutions.
  • E. Constitution Avenue
    Constitution Avenue is a major ceremonial and administrative thoroughfare in Islamabad, Pakistan, lined with key government buildings and national institutions.
  • 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_69d822e5911c8190ba589f957dbd9ba7 completed April 9, 2026, 10:06 p.m.
NER Named-entity recognition batch_69dec72ea9348190817efcdaa973d7f7 completed April 14, 2026, 11:01 p.m.
NED1 Entity disambiguation (via context triple) batch_69ff01d44044819088c86f46b02404ed completed May 9, 2026, 9:43 a.m.
Created at: April 10, 2026, 1:29 a.m.