Triple
T20869454
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | + (album) |
E513851
|
entity |
| Predicate | track |
P17929
|
FINISHED |
| Object | Drunk |
—
|
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: Drunk | Statement: [+ (album), track, Drunk]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Drunk Context triple: [+ (album), track, Drunk]
-
A.
Drunk
Drunk is a 2010 Danish drama film directed by Thomas Vinterberg that explores the consequences of a group of teachers testing a theory that maintaining a constant level of alcohol in their blood will improve their lives.
-
B.
Drunk
chosen
"Drunk" is a song best known as a hit single by English singer-songwriter Ed Sheeran from his debut studio album, +.
-
C.
Still Drunk
"Still Drunk" is a song by the band Afraid of Ghosts, likely reflecting their indie or alternative rock style and introspective themes.
-
D.
Day Drunk
"Day Drunk" is a pop song co-written by American songwriter Lindy Robbins, known for its catchy melody and relatable, laid-back lyrics about daytime drinking.
-
E.
Driven to Drink
Driven to Drink is a stand-up comedy album by comedian Paul F. Tompkins, showcasing his witty, storytelling-based observational humor.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (2 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_69e0b4f675cc8190b4e745225b62eb66 |
completed | April 16, 2026, 10:07 a.m. |
| NER | Named-entity recognition | batch_69e6c4624ce48190b85e5bb24cf0a305 |
completed | April 21, 2026, 12:27 a.m. |
Created at: April 16, 2026, 12:45 p.m.