Triple
T3779
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Lincoln Laboratory |
E71
|
entity |
| Predicate | collaboratesWith |
P37
|
FINISHED |
| Object | U.S. military services |
—
|
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: U.S. military services | Statement: [Lincoln Laboratory, collaboratesWith, U.S. military services]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: collaboratesWith Context triple: [Lincoln Laboratory, collaboratesWith, U.S. military services]
-
A.
relatedTo
chosen
Indicates a general, non-specific relationship or association exists between two entities.
-
B.
grantedTo
Indicates that a right, permission, or resource has been formally given or assigned by one party to another.
-
C.
commissionedBy
Indicates that one entity has been formally requested, authorized, or hired by another entity to create, perform, or carry out something.
-
D.
complements
Indicates that one entity enhances, completes, or improves another by providing qualities or functions that fit well together.
-
E.
proposes
Indicates that one entity formally suggests or puts forward an idea, plan, or course of action to another entity for consideration or approval.
- 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_69a238d6b47881909e68288aed2fd858 |
completed | Feb. 28, 2026, 12:37 a.m. |
| NER | Named-entity recognition | batch_69a23bcc8eb48190b897cc331563980a |
completed | Feb. 28, 2026, 12:50 a.m. |
| PD | Predicate disambiguation | batch_69a23994309081909ff3e869deef2156 |
completed | Feb. 28, 2026, 12:40 a.m. |
Created at: Feb. 28, 2026, 12:40 a.m.