Triple

T5362521
Position Surface form Disambiguated ID Type / Status
Subject North Block E103054 entity
Predicate faces P1699 FINISHED
Object South Block E105125 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: South Block | Statement: [North Block, faces, South Block]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: South Block
Context triple: [North Block, faces, South Block]
  • A. South Block chosen
    South Block is a prominent government building in New Delhi that houses key offices of India’s central administration, including the Prime Minister’s Office and the Ministries of Defence and External Affairs.
  • B. North Block
    North Block is a prominent government building in New Delhi that houses key ministries of the Government of India, including the Ministry of Finance and the Ministry of Home Affairs.
  • C. Crown Square
    Crown Square is the main courtyard and ceremonial heart of Edinburgh Castle, surrounded by some of its most important historic buildings.
  • D. Waverley Square
    Waverley Square is a commercial and transit-oriented neighborhood in Belmont, Massachusetts, known for its local shops and MBTA commuter rail station.
  • E. Regent Square
    Regent Square is a leafy, historic Pittsburgh neighborhood known for its walkable streets, classic homes, and proximity to Frick Park.
  • 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_69bd43daa3e4819090b59d127db70e57 completed March 20, 2026, 12:55 p.m.
NER Named-entity recognition batch_69bd865b9b808190a1e8c283ba28d645 completed March 20, 2026, 5:39 p.m.
NED1 Entity disambiguation (via context triple) batch_69bf29205b188190a19cacbded1d5418 completed March 21, 2026, 11:26 p.m.
Created at: March 20, 2026, 2:02 p.m.