Triple
T10583471
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Before the Devil Knows You're Dead |
E249792
|
entity |
| Predicate | productionCompany |
P490
|
FINISHED |
| Object | Linsefilm |
E773962
|
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: Linsefilm | Statement: [Before the Devil Knows You're Dead, productionCompany, Linsefilm]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Linsefilm Context triple: [Before the Devil Knows You're Dead, productionCompany, Linsefilm]
-
A.
Celluloid
Celluloid is a lightweight, open-source media player for Linux that provides a simple GTK-based interface for the MPV playback engine.
-
B.
Kino-Eye
Kino-Eye was a Soviet avant-garde film movement led by Dziga Vertov that championed documentary realism and the use of the camera as a tool to reveal truth through innovative editing and non-fiction imagery.
-
C.
Planfilm
Planfilm is a film distribution company known for handling releases such as Orson Welles' documentary-style film "F for Fake."
-
D.
Lenfilm
chosen
Lenfilm is one of Russia’s oldest and most prominent film studios, based in Saint Petersburg and known for producing many classic Soviet-era movies.
-
E.
Holga
Holga is a fictional character appearing in the post-apocalyptic virtual reality shooter game "After the Fall."
- 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_69d381c9d3d48190a29ee491e1696a0e |
completed | April 6, 2026, 9:50 a.m. |
| NER | Named-entity recognition | batch_69d52767d2e0819099511e29e254bc34 |
completed | April 7, 2026, 3:48 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69d94b820330819090ed70c6ba29b5c2 |
completed | April 10, 2026, 7:12 p.m. |
Created at: April 6, 2026, 12:39 p.m.