Triple
T18016575
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | RetinaNet |
E431010
|
entity |
| Predicate | lossFunctionProperty |
P43002
|
FINISHED |
| Object | down-weights easy examples |
—
|
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: down-weights easy examples | Statement: [RetinaNet, lossFunctionProperty, down-weights easy examples]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: lossFunctionProperty Context triple: [RetinaNet, lossFunctionProperty, down-weights easy examples]
-
A.
usesLossFunction
Indicates that one entity employs a particular loss function as part of its optimization or learning process.
-
B.
lossType
chosen
Indicates the specific category or nature of a loss associated with an entity or event.
-
C.
performanceFunction
Indicates a relationship where a function or mapping quantifies or evaluates the performance level of an entity, action, or system.
-
D.
lostFunction
Indicates that an entity no longer possesses or can perform a function or capability it previously had.
-
E.
activationFunction
Indicates the specific mathematical transformation applied to a neuron's input to produce its output in a computational or neural model.
- 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_69d8b904530081908bf341d842464856 |
completed | April 10, 2026, 8:47 a.m. |
| NER | Named-entity recognition | batch_69e4b9be5d0c819097e006f32d98753a |
completed | April 19, 2026, 11:17 a.m. |
| PD | Predicate disambiguation | batch_69e3f904b8048190add43883cd7cb191 |
completed | April 18, 2026, 9:35 p.m. |
Created at: April 10, 2026, 10:24 a.m.