Triple
T21193593
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Su Tseng-chang |
E522269
|
entity |
| Predicate | precededBy |
P97
|
FINISHED |
| Object | Frank Hsieh |
—
|
NE NERFINISHED |
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: Frank Hsieh | Statement: [Su Tseng-chang, precededBy, Frank Hsieh]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Frank Hsieh Context triple: [Su Tseng-chang, precededBy, Frank Hsieh]
-
A.
Frank Hsieh
chosen
Frank Hsieh is a Taiwanese politician and former mayor of Kaohsiung who later served as Premier and became a prominent figure in the Democratic Progressive Party.
-
B.
Yao-Hung Hubert Tsai
Yao-Hung Hubert Tsai is a computer scientist and researcher known for his work in self-supervised learning and speech representation, including contributions to the HuBERT model.
-
C.
Chien Wen-pin
Chien Wen-pin is a Taiwanese conductor known for his work with major orchestras in Asia and Europe.
-
D.
Yang Cho-cheng
Yang Cho-cheng was a prominent Taiwanese architect known for designing many of Taipei’s landmark public buildings and shaping the city’s modern architectural landscape.
-
E.
Tsung-Yi Lin
Tsung-Yi Lin is a computer vision researcher known for influential work in object detection and instance segmentation, including co-developing the RetinaNet detector and the COCO dataset.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (2 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_69e0b51061388190aa03f19700d3ef04 |
completed | April 16, 2026, 10:08 a.m. |
| NER | Named-entity recognition | batch_69e73339aaa081909d9009c58c386422 |
completed | April 21, 2026, 8:20 a.m. |
Created at: April 16, 2026, 3:08 p.m.