Triple
T7846130
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Form W-3 |
E181926
|
entity |
| Predicate | reportsOnBehalfOf |
P4131
|
FINISHED |
| Object | employees |
—
|
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: employees | Statement: [Form W-3, reportsOnBehalfOf, employees]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: reportsOnBehalfOf Context triple: [Form W-3, reportsOnBehalfOf, employees]
-
A.
appointedOnBehalfOf
Indicates that an appointment or designation is made by one party acting in representation of, or for the benefit of, another party.
-
B.
builtOnBehalfOf
Indicates that something was constructed or created by one party acting as a representative or agent for another party.
-
C.
reportsThrough
Indicates that one entity formally communicates and is accountable to another entity through an established reporting or supervisory chain.
-
D.
actsOnBehalfOf
chosen
Indicates that one entity performs actions or makes decisions as a representative or proxy for another entity.
-
E.
reportedOn
Indicates that one entity has provided an account, description, or coverage of another entity, typically as news, documentation, or an official report.
- 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_69ca8285d6488190a95d4c02d7354b53 |
completed | March 30, 2026, 2:02 p.m. |
| NER | Named-entity recognition | batch_69cb163f5da08190b7340ef4cc01cd19 |
completed | March 31, 2026, 12:33 a.m. |
| PD | Predicate disambiguation | batch_69cae92180f88190ae3d44c3de7adc93 |
completed | March 30, 2026, 9:20 p.m. |
Created at: March 30, 2026, 4:49 p.m.