Triple
T5213762
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Dunkirk pocket |
E117697
|
entity |
| Predicate | encirclingForce |
P24785
|
FINISHED |
| Object | German forces |
—
|
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: German forces | Statement: [Dunkirk pocket, encirclingForce, German forces]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: encirclingForce Context triple: [Dunkirk pocket, encirclingForce, German forces]
-
A.
encirclementOperation
Indicates a military action in which forces maneuver to surround an opposing force, cutting off its escape routes and lines of support.
-
B.
besiegingForce
chosen
Indicates a military group that is surrounding and attacking a target location or force in an attempt to capture or subdue it.
-
C.
garrisonedForce
Indicates that a military force is stationed in and occupies a specific location, typically for defense or control.
-
D.
engagedForces
Indicates that one force has actively committed or deployed its military units against another force in combat or operational interaction.
-
E.
capturedForce
Indicates that one entity has taken control of another entity or group through force, restraint, or coercive action.
- 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_69bd4464ba3c8190bc16b2ebbe42ddb0 |
completed | March 20, 2026, 12:58 p.m. |
| NER | Named-entity recognition | batch_69bd7a911d40819086621537274dc0f0 |
completed | March 20, 2026, 4:49 p.m. |
| PD | Predicate disambiguation | batch_69bd77bb4e8c819094b5ac7cf61512f9 |
completed | March 20, 2026, 4:37 p.m. |
Created at: March 20, 2026, 1:47 p.m.