Triple
T31282417
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Form 1040-A |
E797705
|
entity |
| Predicate | didNotAllow |
P12540
|
FINISHED |
| Object | itemized deductions on Schedule A |
—
|
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: itemized deductions on Schedule A | Statement: [Form 1040-A, didNotAllow, itemized deductions on Schedule A]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: didNotAllow Context triple: [Form 1040-A, didNotAllow, itemized deductions on Schedule A]
-
A.
doesNotAllow
chosen
Indicates that one entity prevents, forbids, or does not permit another entity from performing a specific action or being in a certain state.
-
B.
deniedAs
Indicates that one entity has refused, rejected, or not granted a request, claim, or association involving another entity.
-
C.
didNotOverrule
Indicates that one entity, typically a higher authority or court, chose not to overturn, reverse, or invalidate the decision or ruling made by another entity.
-
D.
typicallyDenies
Indicates that an entity, in most usual or common circumstances, refuses, rejects, or does not grant something to another entity.
-
E.
deniedTo
Indicates that access, permission, or a requested benefit is explicitly refused or withheld from a particular entity.
- 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_69f224def9088190a37034eab3daf57f |
completed | April 29, 2026, 3:33 p.m. |
| NER | Named-entity recognition | batch_69f69edbb7648190bd89c57e0932eac1 |
completed | May 3, 2026, 1:03 a.m. |
| PD | Predicate disambiguation | batch_69f69d1bf8cc8190a78dfa5ab00daf3a |
completed | May 3, 2026, 12:55 a.m. |
Created at: April 29, 2026, 9:13 p.m.