Triple

T3381294
Position Surface form Disambiguated ID Type / Status
Subject U6 E71188 entity
Predicate hasStation P35 FINISHED
Object Hallesches Tor E72035 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: Hallesches Tor | Statement: [U6, hasStation, Hallesches Tor]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Hallesches Tor
Context triple: [U6, hasStation, Hallesches Tor]
  • A. Hallesches Tor chosen
    Hallesches Tor is a major Berlin U-Bahn interchange station in the Kreuzberg district, serving as a key hub for multiple subway lines.
  • B. Oranienburger Tor
    Oranienburger Tor is an underground station on Berlin’s U-Bahn network located near the historic Oranienburger Tor area in the central district of Mitte.
  • C. Schottentor
    Schottentor is a historic former city gate area in Vienna that now serves as a major public transport hub and landmark at the edge of the Innere Stadt.
  • D. Kröpeliner Tor
    Kröpeliner Tor is a historic medieval city gate and prominent architectural landmark in the German city of Rostock.
  • E. Sandauer Tor
    Sandauer Tor is a historic city gate in Landsberg am Lech, Germany, forming part of the town’s preserved medieval fortifications.
  • 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_69ad85a7f80c8190a05e43013f298942 completed March 8, 2026, 2:20 p.m.
NER Named-entity recognition batch_69adb5e9af608190bfb228ef99a87bb7 completed March 8, 2026, 5:46 p.m.
NED1 Entity disambiguation (via context triple) batch_69b34bc3f13c81909bec375bd080b7f3 completed March 12, 2026, 11:27 p.m.
Created at: March 8, 2026, 3:14 p.m.