Triple
T589239
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Japanese Special Naval Landing Forces |
E17229
|
entity |
| Predicate | allegedToHaveCommitted |
P9856
|
FINISHED |
| Object | war crimes |
—
|
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: war crimes | Statement: [Japanese Special Naval Landing Forces, allegedToHaveCommitted, war crimes]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: allegedToHaveCommitted Context triple: [Japanese Special Naval Landing Forces, allegedToHaveCommitted, war crimes]
-
A.
accusedOf
chosen
Indicates that one entity has formally alleged or claimed that another entity committed a specific wrongdoing or offense.
-
B.
guiltyOf
Indicates that an entity has been judged or determined to have committed a particular offense, crime, or wrongful act.
-
C.
convictedOf
Indicates that a person or entity has been found guilty of committing a specified offense or crime through a formal legal process.
-
D.
committedCrime
Indicates that an entity has carried out or been responsible for a criminal act or offense.
-
E.
confessedTo
Indicates that one entity admitted guilt or revealed the truth about an action, wrongdoing, or secret to another 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_69a49379d09c8190ac7e00b24e2810b1 |
completed | March 1, 2026, 7:28 p.m. |
| NER | Named-entity recognition | batch_69a49bb775fc819085b968f8615dca59 |
completed | March 1, 2026, 8:04 p.m. |
| PD | Predicate disambiguation | batch_69a494cc13988190892ca10bd7ae9f09 |
completed | March 1, 2026, 7:34 p.m. |
Created at: March 1, 2026, 7:33 p.m.