Triple
T28047923
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Dr. Tyrone Berger |
E708732
|
entity |
| Predicate | helpsCharacterWith |
P44218
|
FINISHED |
| Object | guilt |
—
|
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: guilt | Statement: [Dr. Tyrone Berger, helpsCharacterWith, guilt]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: helpsCharacterWith Context triple: [Dr. Tyrone Berger, helpsCharacterWith, guilt]
-
A.
helpsProtagonistWith
Indicates that one entity assists the protagonist in performing, achieving, or dealing with something specified by the other entity.
-
B.
helpsCharacterConfront
chosen
Indicates that one character actively supports or enables another character in facing and dealing with a difficult issue, fear, or challenge.
-
C.
helpsCharacterType
Indicates that one character type provides assistance or support to another character type.
-
D.
collaboratesWithCharacter
Indicates that one character works together with another character toward a shared goal or activity.
-
E.
supportCharacter
Indicates that one entity provides assistance, backing, or reinforcement to another entity in performing an action or fulfilling a role.
- 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_69ef9b6df9f48190bbb971d02cbe1b65 |
completed | April 27, 2026, 5:22 p.m. |
| NER | Named-entity recognition | batch_69f63fd79e4c8190af9263b679e5ff07 |
completed | May 2, 2026, 6:17 p.m. |
| PD | Predicate disambiguation | batch_69f63c6a8474819091b8c6fe98e3862d |
completed | May 2, 2026, 6:03 p.m. |
Created at: April 27, 2026, 8:31 p.m.