Triple
T18357390
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Run for Your Life |
E439829
|
entity |
| Predicate | containsLyricReference |
P85363
|
FINISHED |
| Object | I’d rather see you dead, little girl |
—
|
LITERAL 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: I’d rather see you dead, little girl | Statement: [Run for Your Life, containsLyricReference, I’d rather see you dead, little girl]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: containsLyricReference Context triple: [Run for Your Life, containsLyricReference, I’d rather see you dead, little girl]
-
A.
hasLyricsMentioning
Indicates that the referenced lyrics explicitly mention or refer to the specified entity.
-
B.
hasLyricsIn
Indicates that the lyrics of a work are written or available in a specified language.
-
C.
lyricReference
Indicates that one entity (typically a work or text) contains or makes a reference to the lyrics of another entity.
-
D.
containsSongReferencing
Indicates that one entity (such as a work, collection, or playlist) includes at least one song that makes an explicit reference to another specified entity.
-
E.
hasLyricPhrase
chosen
Indicates that one entity (typically a song or musical work) contains or is associated with a specific lyric phrase as part of its textual content.
- F. None of above.
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_69d8b918221c8190a9f7b563d64ac677 |
completed | April 10, 2026, 8:47 a.m. |
| NER | Named-entity recognition | batch_69e516d776cc8190937d7e1d42a36a3b |
completed | April 19, 2026, 5:54 p.m. |
| PD | Predicate disambiguation | batch_69e44fed3fdc81908f4ed6a81db42416 |
completed | April 19, 2026, 3:45 a.m. |
Created at: April 10, 2026, 10:37 a.m.