Triple
T4397728
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Dependent and Disability Pension Act |
E99532
|
entity |
| Predicate | replacedSystemBasedOn |
P55858
|
FINISHED |
| Object | service-related injuries only |
—
|
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: service-related injuries only | Statement: [Dependent and Disability Pension Act, replacedSystemBasedOn, service-related injuries only]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: replacedSystemBasedOn Context triple: [Dependent and Disability Pension Act, replacedSystemBasedOn, service-related injuries only]
-
A.
replacedSystemUsedUntil
Indicates that one system was used up until it was replaced by another system.
-
B.
replacedInSystemVersion
Indicates that one entity has been superseded or substituted by another in a specific version of a system.
-
C.
replacedSubsystem
Indicates that one subsystem has been substituted or exchanged for another, taking over its role or function.
-
D.
rebootOf
Indicates that one work is a new version that restarts or reimagines the story, continuity, or franchise of another earlier work.
-
E.
partlyReplacedBy
Indicates that one entity has been superseded or substituted in part, but not entirely, by another entity.
- F. None of above. chosen
Provenance (4 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_69b345506b408190b0e3dee616738a7d |
completed | March 12, 2026, 10:59 p.m. |
| NER | Named-entity recognition | batch_69b352adce588190b9e6ed53458aa1e1 |
completed | March 12, 2026, 11:56 p.m. |
| PD | Predicate disambiguation | batch_69b34f597998819092477efdedb51427 |
completed | March 12, 2026, 11:42 p.m. |
| PDg | Predicate description generation | batch_69b34ff654308190b9717526120d80d3 |
completed | March 12, 2026, 11:44 p.m. |
Created at: March 12, 2026, 11:20 p.m.