Triple

T23879141

Position	Surface form	Disambiguated ID	Type / Status
Subject	Gran Sasso raid	`E600149`	entity
Predicate	guardingForce	`P97703`	FINISHED
Object	Italian Carabinieri detachment The Italian Carabinieri detachment was a unit of Italy’s national military police responsible for security and law enforcement duties, including guarding high-profile prisoners and strategic sites during World War II.	`E229578`	NE FINISHED

How this triple was built (3 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: Italian Carabinieri detachment | Statement: [Gran Sasso raid, guardingForce, Italian Carabinieri detachment]

NEDg Description generation gpt-5.1

Instruction

Generate a one-sentence description of the target entity. 
You are given a context triple in the form (subject, predicate, object), where the object is the target entity. 
# Instructions
Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. 
Avoid repeating the information from the triple, unless really essential.
# Response Format
Return only the sentence: "Description: [one-sentence description of the target entity]"

Input

Entity: Italian Carabinieri detachment
Triple: [Gran Sasso raid, guardingForce, Italian Carabinieri detachment]

Generated description

The Italian Carabinieri detachment was a unit of Italy’s national military police responsible for security and law enforcement duties, including guarding high-profile prisoners and strategic sites during World War II.

PD Predicate disambiguation gpt-5-mini-2025-08-07

Target predicate: guardingForce
Context triple: [Gran Sasso raid, guardingForce, Italian Carabinieri detachment]

A. garrisonedForce
Indicates that a military force is stationed in and occupies a specific location, typically for defense or control.
B. besiegingForce
Indicates a military group that is surrounding and attacking a target location or force in an attempt to capture or subdue it.
C. protectingPowerFor
Indicates that one entity serves as a protecting power, safeguarding the interests or rights of another entity, typically in a diplomatic or conflict-related context.
D. encircledForce
Indicates that a military force is surrounded by enemy units, cutting off its movement or escape routes.
E. guard chosen
Indicates that one entity protects, watches over, or defends another entity or resource from harm, intrusion, or unauthorized access.
F. None of above.

Provenance (6 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_69e295318e148190b9979d8fc02e168f`	completed	April 17, 2026, 8:16 p.m.
NER	Named-entity recognition	`batch_69f1cc03fdcc8190ad77155ae7e6eda5`	completed	April 29, 2026, 9:14 a.m.
NED1	Entity disambiguation (via context triple)	`batch_6a0f69c8c6808190b14117a0d67aba0e`	completed	May 21, 2026, 8:23 p.m.
NEDg	Description generation	`batch_6a0f6d4399e481908e08d9dc8dd5e139`	completed	May 21, 2026, 8:38 p.m.
NED2	Entity disambiguation (via description)	`batch_6a0f6e38cf648190b30122be93c70685`	completed	May 21, 2026, 8:42 p.m.
PD	Predicate disambiguation	`batch_69f1614a65a88190bde1efb368a151e4`	completed	April 29, 2026, 1:39 a.m.

Created at: April 17, 2026, 8:23 p.m.