Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele

Itọkasi Ede Adayeba ati Idanwo Entailment

Learn NLI’s entailment, contradiction, and neutral labels, benchmark datasets, transformer inputs, and evidence-checking limits.

Jẹmọ itọsọna awọn ọnaAdayeba Ede Inference Ati Entailment
Ibeere 1 ti 8

Kini awọn aami idiwọn mẹta ni itọka ede adayeba?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. Kini awọn aami idiwọn mẹta ni itọka ede adayeba?
    • Entailment, ilodi, didoju
    • Rere, odi, didoju
    • Otitọ, eke, aimọ
    • Koko-ọrọ, ọrọ-ọrọ, nkan
  2. Ni NLI, kini 'entailment' tumọ si?
    • Awọn gbolohun ọrọ meji ko ni ibatan
    • Iroran naa gbọdọ jẹ otitọ fun ipilẹ-ile
    • Awọn ilewq gbọdọ jẹ eke fi fun awọn ayika ile
    • Awọn gbolohun ọrọ jẹ aami girama
  3. Ti a fun ni ipilẹ 'Ọkunrin kan n ṣe gita lori ipele,' idawọle wo ni o jẹ ilodi?
    • Eniyan n ṣe orin
    • Awọn eniyan fẹràn orin naa
    • Ipele ti ṣofo
    • Olorin mu ohun-elo kan mu
  4. What is an annotation artifact in an NLI dataset?
    • A model that reasons perfectly with very little data
    • A dataset where every annotator agrees
    • A premise that is always longer than its hypothesis
    • A spurious cue that lets a model guess labels without reasoning about the premise-hypothesis relation
  5. Awọn ipilẹ data wo ni o lo pupọ lati ṣe ikẹkọ ati ṣe iṣiro NLI?
    • SNLI ati MultiNLI
    • ImageNet ati CIFAR
    • MNIST ati SVHN
    • WikiText ati Penn Treebank
  6. How can NLI support a check of a generated claim against source evidence?
    • Measure the generated claim’s length
    • Classify whether the evidence entails, contradicts, or leaves the claim neutral
    • Count unique words only
    • Translate the claim into another language
  7. In NLI, what does the neutral label mean?
    • The hypothesis is guaranteed true
    • The hypothesis is guaranteed false
    • The premise neither entails nor contradicts the hypothesis
    • The two sentences are grammatically identical
  8. How does a common transformer NLI classifier present a sentence pair to the model?
    • Encode only the hypothesis and discard the premise
    • Convert both sentences into audio first
    • Train a separate model for every word
    • Concatenate the premise and hypothesis with a separator token