Triple

T18204950
Position Surface form Disambiguated ID Type / Status
Subject Longformer E435878 entity
Predicate relatedTo P37 FINISHED
Object Reformer
Reformer is a Transformer-based neural network architecture that uses locality-sensitive hashing and reversible layers to achieve more memory- and computation-efficient attention on long sequences.
 E1312467 NE FINISHED

How this triple was built (4 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: Reformer | Statement: [Longformer, relatedTo, Reformer]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07 
Target entity: Reformer
Context triple: [Longformer, relatedTo, Reformer]
  • A. Fiser
    Fiser is a surname variant of Fischer, commonly associated with Central or Eastern European origins.
  • B. Schröder
    Schröder is a common German surname borne by numerous notable figures in politics, sports, and the arts.
  • C. Bryc
    Bryc is an alternative spelling of the given name Bryce, typically used as a modern or stylistic variant.
  • D. Bridgman
    Bridgman is a surname most notably associated with American physicist and Nobel laureate Percy Williams Bridgman, a pioneer in high-pressure physics.
  • E. Fichtner
    Fichtner is a German-origin surname most notably associated with American character actor William Fichtner, known for his roles in film and television.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
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: Reformer
Triple: [Longformer, relatedTo, Reformer]
Generated description
Reformer is a Transformer-based neural network architecture that uses locality-sensitive hashing and reversible layers to achieve more memory- and computation-efficient attention on long sequences.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07 
Target entity: Reformer
Target entity description: Reformer is a Transformer-based neural network architecture that uses locality-sensitive hashing and reversible layers to achieve more memory- and computation-efficient attention on long sequences.
  • A. Fiser
    Fiser is a surname variant of Fischer, commonly associated with Central or Eastern European origins.
  • B. Schröder
    Schröder is a common German surname borne by numerous notable figures in politics, sports, and the arts.
  • C. Bryc
    Bryc is an alternative spelling of the given name Bryce, typically used as a modern or stylistic variant.
  • D. Bridgman
    Bridgman is a surname most notably associated with American physicist and Nobel laureate Percy Williams Bridgman, a pioneer in high-pressure physics.
  • E. Fichtner
    Fichtner is a German-origin surname most notably associated with American character actor William Fichtner, known for his roles in film and television.
  • F. None of above. chosen

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_69d8b90dba6481908e119eb9aa4ca0cb completed April 10, 2026, 8:47 a.m.
NER Named-entity recognition batch_69e4e222831081908f7d5500424e3acb completed April 19, 2026, 2:09 p.m.
NED1 Entity disambiguation (via context triple) batch_6a039f0e52108190913cc5c667619d89 completed May 12, 2026, 9:43 p.m.
NEDg Description generation batch_6a039fdd9c4c819083b450657d0ece43 completed May 12, 2026, 9:47 p.m.
NED2 Entity disambiguation (via description) batch_6a03a0d6de8c8190b1f94c7de0856143 completed May 12, 2026, 9:51 p.m.
Created at: April 10, 2026, 10:32 a.m.