Triple

T6796286
Position Surface form Disambiguated ID Type / Status
Subject Small Gods E156060 entity
Predicate featuresCharacter P626 FINISHED
Object Vorbis
Vorbis is a fanatical and manipulative high priest and inquisitor who serves as the primary antagonist in Terry Pratchett’s Discworld novel "Small Gods."
E620094 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: Vorbis | Statement: [Small Gods, featuresCharacter, Vorbis]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Vorbis
Context triple: [Small Gods, featuresCharacter, Vorbis]
  • A. Speex
    Speex is an open-source, patent-free audio compression codec designed primarily for efficient, high-quality speech encoding.
  • B. OGG
    OGG is the IATA airport code for Kahului Airport, the primary commercial airport serving the island of Maui in Hawaii.
  • C. MPEG-1 Layer II audio
    MPEG-1 Layer II audio is a lossy audio compression format widely used in digital broadcasting and early digital media for efficient, good-quality stereo sound.
  • D. Matroska
    Matroska is an open standard, extensible multimedia container format best known for its MKV files that can hold multiple video, audio, subtitle, and metadata streams in a single package.
  • E. FLAC
    FLAC (Free Lossless Audio Codec) is an open-source audio compression format that reduces file size without any loss in sound quality, commonly used for high-fidelity music archiving and playback.
  • 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: Vorbis
Triple: [Small Gods, featuresCharacter, Vorbis]
Generated description
Vorbis is a fanatical and manipulative high priest and inquisitor who serves as the primary antagonist in Terry Pratchett’s Discworld novel "Small Gods."
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07
Target entity: Vorbis
Target entity description: Vorbis is a fanatical and manipulative high priest and inquisitor who serves as the primary antagonist in Terry Pratchett’s Discworld novel "Small Gods."
  • A. Speex
    Speex is an open-source, patent-free audio compression codec designed primarily for efficient, high-quality speech encoding.
  • B. Cinepak
    Cinepak is an early video compression codec widely used in the 1990s for CD-ROM and multimedia applications due to its ability to play smoothly on low-powered hardware.
  • C. OGG
    OGG is the IATA airport code for Kahului Airport, the primary commercial airport serving the island of Maui in Hawaii.
  • D. MPEG-1 Layer II audio
    MPEG-1 Layer II audio is a lossy audio compression format widely used in digital broadcasting and early digital media for efficient, good-quality stereo sound.
  • E. Matroska
    Matroska is an open standard, extensible multimedia container format best known for its MKV files that can hold multiple video, audio, subtitle, and metadata streams in a single package.
  • 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_69c6881844448190a65822d9b39d7f88 completed March 27, 2026, 1:37 p.m.
NER Named-entity recognition batch_69c6d2c872f0819092f21e8fdbecc667 completed March 27, 2026, 6:56 p.m.
NED1 Entity disambiguation (via context triple) batch_69c71a95987c8190ae8ed5840a6744f0 completed March 28, 2026, 12:02 a.m.
NEDg Description generation batch_69c71ba85f608190b372a4dfe2cdc31c completed March 28, 2026, 12:07 a.m.
NED2 Entity disambiguation (via description) batch_69c71c91e08c81908be81efc2087464a completed March 28, 2026, 12:10 a.m.
Created at: March 27, 2026, 2:15 p.m.