Triple
T1603254
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Battle of Tora Bora |
E34441
|
entity |
| Predicate | opponentStrength |
P30658
|
FINISHED |
| Object | hundreds of al-Qaeda fighters |
—
|
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: hundreds of al-Qaeda fighters | Statement: [Battle of Tora Bora, opponentStrength, hundreds of al-Qaeda fighters]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: opponentStrength Context triple: [Battle of Tora Bora, opponentStrength, hundreds of al-Qaeda fighters]
-
A.
opponentObjective
Indicates that one entity is the target or goal that an opposing entity aims to counter, defeat, or prevent from succeeding.
-
B.
keyOpponents
Indicates that the subject has primary or most significant opponents identified by the object.
-
C.
battleOpponent
Indicates that two entities are engaged in or designated as opponents in a battle or combat scenario.
-
D.
opponentState
Indicates the condition or status that an opposing party or competitor is currently in within a given context or interaction.
-
E.
decidingOpponent
Indicates that one entity is the opponent or counterpart responsible for making a decision that affects another entity.
- F. None of above. chosen
Provenance (4 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_69a885fea6a481909fe83ba6441f1774 |
completed | March 4, 2026, 7:20 p.m. |
| NER | Named-entity recognition | batch_69a95b02cd448190be8e3db9a5a7bac0 |
completed | March 5, 2026, 10:29 a.m. |
| PD | Predicate disambiguation | batch_69a907c1cad08190b9728dd557f39aa0 |
completed | March 5, 2026, 4:34 a.m. |
| PDg | Predicate description generation | batch_69a95aada3f881909053363c01de8b57 |
completed | March 5, 2026, 10:29 a.m. |
Created at: March 4, 2026, 7:28 p.m.