Triple
T32021637
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | North White Plains station |
E817704
|
entity |
| Predicate | followingStationOnHarlemLine |
P195594
|
FINISHED |
| Object | Valhalla station |
—
|
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: Valhalla station | Statement: [North White Plains station, followingStationOnHarlemLine, Valhalla station]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: followingStationOnHarlemLine Context triple: [North White Plains station, followingStationOnHarlemLine, Valhalla station]
-
A.
precedingStationOnHarlemLine
Indicates that one station is immediately before another station in travel order along the Harlem Line.
-
B.
adjacentStationOnHarlemLine
Indicates that one station is directly next to another station along the Harlem Line, with no other station in between.
-
C.
followingStationOnYellowLine
Indicates that one station is the next station in sequence after another on the Yellow Line.
-
D.
followingStationOnBlueLine
Indicates that one station is the next consecutive stop after another station on the Blue Line route.
-
E.
followingStationOnRedLine
Indicates that one station is the next station immediately following another station along the Red Line route.
- F. None of above. chosen
Provenance (4 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_69f348fb04e4819081f4eab040ed7959 |
completed | April 30, 2026, 12:20 p.m. |
| NER | Named-entity recognition | batch_69fddac4e2f48190a9301d3422658b29 |
completed | May 8, 2026, 12:44 p.m. |
| PD | Predicate disambiguation | batch_69fdda06969c8190b5d033964ea2a690 |
completed | May 8, 2026, 12:41 p.m. |
| PDg | Predicate description generation | batch_69fddac42c4081908568649058e86458 |
completed | May 8, 2026, 12:44 p.m. |
Created at: May 1, 2026, 12:16 a.m.