Triple

T26673181

Position	Surface form	Disambiguated ID	Type / Status
Subject	The Life Scientific	`E672388`	entity
Predicate	producer	`P490`	FINISHED
Object	BBC Radio Science Unit The BBC Radio Science Unit is a specialist production team within the BBC that creates science-focused radio programmes and podcasts for audiences in the UK and beyond.	`E1734220`	NE 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: BBC Radio Science Unit | Statement: [The Life Scientific, producer, BBC Radio Science Unit]

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: BBC Radio Science Unit
Triple: [The Life Scientific, producer, BBC Radio Science Unit]

Generated description

The BBC Radio Science Unit is a specialist production team within the BBC that creates science-focused radio programmes and podcasts for audiences in the UK and beyond.

Provenance (5 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_69eecda00a9c8190b2691f4d89db03b6`	completed	April 27, 2026, 2:44 a.m.
NER	Named-entity recognition	`batch_69f6170129f88190a8d34212c7e030a4`	completed	May 2, 2026, 3:23 p.m.
NED1	Entity disambiguation (via context triple)	`batch_6a11ec52e70c819080f9767f7a481b09`	completed	May 23, 2026, 6:05 p.m.
NEDg	Description generation	`batch_6a11ed2cc62c8190b582f46a4b2ca426`	completed	May 23, 2026, 6:08 p.m.
NED2	Entity disambiguation (via description)	`batch_6a11edac59388190bfa4e3e288b7932e`	completed	May 23, 2026, 6:10 p.m.

Created at: April 27, 2026, 3:14 a.m.