Triple

T22411996
Position Surface form Disambiguated ID Type / Status
Subject ImageNet Classification with Deep Convolutional Neural Networks E554013 entity
Predicate improvementOverStateOfTheArtTop5Error P148054 FINISHED
Object more than 10 percentage points 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: more than 10 percentage points | Statement: [ImageNet Classification with Deep Convolutional Neural Networks, improvementOverStateOfTheArtTop5Error, more than 10 percentage points]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: improvementOverStateOfTheArtTop5Error
Context triple: [ImageNet Classification with Deep Convolutional Neural Networks, improvementOverStateOfTheArtTop5Error, more than 10 percentage points]
  • A. top5ErrorRate
    Indicates the proportion of instances where the correct answer is not among the top five predicted results.
  • B. bestF1Result
    Indicates that one result in a set has the highest F1 score (harmonic mean of precision and recall) compared to all other results.
  • C. inceptionApproximation
    Indicates an approximate or estimated starting point or origin of something, rather than an exact inception time.
  • D. usesNeuralNetworks
    Indicates that one entity employs neural network models or techniques as part of its functioning, processing, or decision-making.
  • E. pretrainedOn
    Indicates that a model has been trained in advance using a specified dataset or data source before being applied to downstream tasks.
  • F. None of above. chosen

Provenance (4 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_69e11e4e6ce8819085a1e06d886bf21c completed April 16, 2026, 5:37 p.m.
NER Named-entity recognition batch_69f15943dd84819099e77563da470594 completed April 29, 2026, 1:05 a.m.
PD Predicate disambiguation batch_69e8989495bc81909d2699fce5992e28 completed April 22, 2026, 9:44 a.m.
PDg Predicate description generation batch_69e8aa39e3388190b659d59948ebf3e6 completed April 22, 2026, 11 a.m.
Created at: April 16, 2026, 8:46 p.m.