Triple
T3624150
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Candice Bergen |
E76795
|
entity |
| Predicate | spouse |
P13
|
FINISHED |
| Object |
Marshall Rose
Marshall Rose is a wealthy New York real estate developer and philanthropist known for his work in high-end property and arts patronage.
|
E372608
|
NE FINISHED |
How this triple was built (4 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: Marshall Rose | Statement: [Candice Bergen, spouse, Marshall Rose]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Marshall Rose Context triple: [Candice Bergen, spouse, Marshall Rose]
-
A.
Andrew Singer
Andrew Singer is a television producer best known for his executive production work on the satirical sketch comedy series "Portlandia."
-
B.
Alexander Rose
Alexander Rose is a historian and author best known for his book "Washington's Spies," which inspired the television series "Turn: Washington's Spies."
-
C.
Richard Marks
Richard Marks was an American film editor known for his work on numerous acclaimed movies, including the culinary drama "Julie & Julia."
-
D.
Andy Lassner
Andy Lassner is a television producer best known for his long-running work on "The Ellen DeGeneres Show" and other major daytime talk shows.
-
E.
Jason Rubin
Jason Rubin is an American video game designer and co-founder of Naughty Dog, best known for his work on the Crash Bandicoot series.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg
Description generation
gpt-5.1
Instruction
Generate a one-sentence description of the target entity. You are given a context triple in the form (subject, predicate, object), where the object is the target entity. # Instructions Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. Avoid repeating the information from the triple, unless really essential. # Response Format Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: Marshall Rose Triple: [Candice Bergen, spouse, Marshall Rose]
Generated description
Marshall Rose is a wealthy New York real estate developer and philanthropist known for his work in high-end property and arts patronage.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: Marshall Rose Target entity description: Marshall Rose is a wealthy New York real estate developer and philanthropist known for his work in high-end property and arts patronage.
-
A.
Andrew Singer
Andrew Singer is a television producer best known for his executive production work on the satirical sketch comedy series "Portlandia."
-
B.
Alexander Rose
Alexander Rose is a historian and author best known for his book "Washington's Spies," which inspired the television series "Turn: Washington's Spies."
-
C.
Richard Marks
Richard Marks was an American film editor known for his work on numerous acclaimed movies, including the culinary drama "Julie & Julia."
-
D.
Andy Lassner
Andy Lassner is a television producer best known for his long-running work on "The Ellen DeGeneres Show" and other major daytime talk shows.
-
E.
Jason Rubin
Jason Rubin is an American video game designer and co-founder of Naughty Dog, best known for his work on the Crash Bandicoot series.
- F. None of above. chosen
Provenance (5 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_69ad85dc03948190b35b7189e4175bcc |
completed | March 8, 2026, 2:21 p.m. |
| NER | Named-entity recognition | batch_69adc2d9845c8190ad65b2471000dfa0 |
completed | March 8, 2026, 6:41 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69b4332260cc8190964a15bfee0a3b61 |
completed | March 13, 2026, 3:54 p.m. |
| NEDg | Description generation | batch_69b435af0f688190ac82dccd487747af |
completed | March 13, 2026, 4:05 p.m. |
| NED2 | Entity disambiguation (via description) | batch_69b43620aee48190953a413ece0d51fa |
completed | March 13, 2026, 4:06 p.m. |
Created at: March 8, 2026, 3:23 p.m.