Triple

T14911028
Position Surface form Disambiguated ID Type / Status
Subject Tukey's biweight E371259 entity
Predicate downweights P9925 FINISHED
Object observations with large residuals LITERAL 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: observations with large residuals | Statement: [Tukey's biweight, downweights, observations with large residuals]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: downweights
Context triple: [Tukey's biweight, downweights, observations with large residuals]
  • A. down
    Indicates that one entity is located at a lower position or level relative to another entity or reference point.
  • B. under
    Indicates that one entity is positioned below or beneath another entity, often implying vertical alignment or coverage.
  • C. isHeavilyWeightedToward
    Indicates that something is strongly biased or disproportionately oriented in favor of one side, option, or aspect over others.
  • D. reduces chosen
    Indicates that one entity causes a decrease in the amount, intensity, degree, or impact of another entity.
  • E. demotes
    Indicates that one entity reduces another entity’s rank, status, or level within a hierarchy or system.
  • F. None of above.

Provenance (3 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_69d85cc7ea3481908228b5acb7d06f12 completed April 10, 2026, 2:13 a.m.
NER Named-entity recognition batch_69ded61c6b9c8190a92934d49b98fe46 completed April 15, 2026, 12:04 a.m.
PD Predicate disambiguation batch_69de9a4a14a88190951bb8f4c60bd37b completed April 14, 2026, 7:49 p.m.
Created at: April 10, 2026, 2:26 a.m.