Triple
T8248507
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Hydra |
E192900
|
entity |
| Predicate | misuseRisk |
P62537
|
FINISHED |
| Object | unauthorized access to systems |
—
|
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: unauthorized access to systems | Statement: [Hydra, misuseRisk, unauthorized access to systems]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: misuseRisk Context triple: [Hydra, misuseRisk, unauthorized access to systems]
-
A.
misuseCanConstitute
Indicates that improper or incorrect use of something can amount to, or be considered as, a particular offense, violation, or condition.
-
B.
riskType
Indicates the category or nature of risk associated with an entity, event, or relationship.
-
C.
riskElement
chosen
Indicates that one entity is a risk-related component, factor, or contributor associated with another entity within a risk context.
-
D.
hasWithdrawalRisk
Indicates that discontinuing or reducing something is associated with a risk of withdrawal effects or adverse reactions.
-
E.
riskLevel
Indicates the degree of potential harm, loss, or adverse outcome associated with a particular situation, action, or 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_69ca82de7b8c81908d8106f8a53cff9b |
completed | March 30, 2026, 2:04 p.m. |
| NER | Named-entity recognition | batch_69cb78c6b3c48190a3ecebf449766124 |
completed | March 31, 2026, 7:33 a.m. |
| PD | Predicate disambiguation | batch_69cb36b437e881909958591357e83b9d |
completed | March 31, 2026, 2:51 a.m. |
Created at: March 30, 2026, 5:48 p.m.