Triple

T7893286
Position Surface form Disambiguated ID Type / Status
Subject Tom Preston-Werner E183287 entity
Predicate workedAt P7 FINISHED
Object Powerset
Powerset was a natural-language search engine startup, later acquired by Microsoft, that focused on enabling more intuitive, semantic web search.
 E697368 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: Powerset | Statement: [Tom Preston-Werner, workedAt, Powerset]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07 
Target entity: Powerset
Context triple: [Tom Preston-Werner, workedAt, Powerset]
  • A. SETS
    SETS is the London Stock Exchange’s central electronic order book system used for automated trading of the most liquid UK securities.
  • B. POWER2
    POWER2 is a second-generation IBM RISC microprocessor architecture designed to deliver high-performance computing, particularly for scientific and technical workloads.
  • C. NSSet
    NSSet is an Objective-C collection class that represents an unordered, unique set of objects, commonly used in Cocoa and Cocoa Touch frameworks.
  • D. POWER1
    POWER1 is IBM’s first-generation 32-bit RISC microprocessor architecture used in early RS/6000 workstations and servers.
  • E. Set
    Set is an ancient Egyptian god associated primarily with chaos, storms, and disorder, often depicted as the adversary of his brother Osiris and the rival of Horus.
  • 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: Powerset
Triple: [Tom Preston-Werner, workedAt, Powerset]
Generated description
Powerset was a natural-language search engine startup, later acquired by Microsoft, that focused on enabling more intuitive, semantic web search.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07 
Target entity: Powerset
Target entity description: Powerset was a natural-language search engine startup, later acquired by Microsoft, that focused on enabling more intuitive, semantic web search.
  • A. SETS
    SETS is the London Stock Exchange’s central electronic order book system used for automated trading of the most liquid UK securities.
  • B. POWER2
    POWER2 is a second-generation IBM RISC microprocessor architecture designed to deliver high-performance computing, particularly for scientific and technical workloads.
  • C. NSSet
    NSSet is an Objective-C collection class that represents an unordered, unique set of objects, commonly used in Cocoa and Cocoa Touch frameworks.
  • D. POWER1
    POWER1 is IBM’s first-generation 32-bit RISC microprocessor architecture used in early RS/6000 workstations and servers.
  • E. Set
    Set is an ancient Egyptian god associated primarily with chaos, storms, and disorder, often depicted as the adversary of his brother Osiris and the rival of Horus.
  • 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_69ca828c474c8190a254d6499871eaff completed March 30, 2026, 2:02 p.m.
NER Named-entity recognition batch_69cb3a008fb88190a039fec40483ab93 completed March 31, 2026, 3:05 a.m.
NED1 Entity disambiguation (via context triple) batch_69cb5ba9c2ac8190b8faf1518390dff4 completed March 31, 2026, 5:29 a.m.
NEDg Description generation batch_69cb5f1e84fc8190b535016cb69405b4 completed March 31, 2026, 5:43 a.m.
NED2 Entity disambiguation (via description) batch_69cb76a214488190b90e5db28511daa0 completed March 31, 2026, 7:24 a.m.
Created at: March 30, 2026, 5 p.m.