quiz bu lëkkale ak tegtal · Diggante Niveau

Laaj ak tontu ci architecture encodeur-decodeur

Xoolaat jumtukaayi encoder-decoder, bàyyi xel ci source-to-target ak jëfandikoo mbind-ci-tekst ci tekki làkk ak liggéeyi làkk.

Yooni njiit yu jëm ciArchitectures Dekodeur
Laaj 1 de 8

Lan mooy liggéey bi gëna am solo ci encoder ci xeetu encoder-decoder?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Lan mooy liggéey bi gëna am solo ci encoder ci xeetu encoder-decoder?
    • Lire liñu dugal ba noppi soppi ko ci ay représentation interne yu am solo
    • Defar génne gi benn jeton benn yoon
    • Raññe njariñu kàndidaa yi ci anam wu mëna am
    • Ngir denc done yiñ tàggat ci model bi
  2. Lu tax dekodeer bi di jëfandikoo maske (causal) bàyyi xel ci boppam?
    • Ngir nëbb dugal bi ci encoder bi
    • Ngir moytu bépp position gis jetons yu ëlëg ci jamonoy generation
    • Ngir kompresse génne gi ci ay token yu néew
    • ngir daaneel ay token ngir yamale
  3. Ban mekaniism mooy boole dekodeer bi ak misaali enkodeer bi?
    • Dàddu
    • Normalisasioŋ ci lots
    • Fexe jàkkarloo
    • Kodage posisioŋ
  4. Ci xeetu architecture yii, yan ci ñoom mooy architecture encodeur-decodeur (toppal-ci-toppal)?
    • GPT-3
    • Bert
    • Word2Vec
    • T5
  5. Lan moo waral jëmmu encoder-decoder bi méngoo bu baax ak tekki làkk?
    • Ndax li ñuy dugal ak li ñuy genne dañu wuute ci toppalante, te fàww ñu xam leen bu baax, ba noppi ñu defar leen ci anam wu bees.
    • Ndax duggal ak genn dañuy nuru benn guddaay
    • Ndax musul soxla defar mbind mu bees
    • Ndax dafay moytu jëfandikoo xel yépp
  6. In the original Transformer architecture, what attention does the encoder use over its input sequence?
    • Causal attention that hides all earlier source tokens
    • Bidirectional self-attention over the full source sequence
    • Cross-attention only to an image sequence
    • No attention mechanism
  7. What training objective does BART use for pretraining?
    • It predicts only the next pixel in an image
    • It classifies each word into a fixed sentiment label
    • It corrupts text and trains a sequence-to-sequence denoising model to reconstruct the original
    • It performs speech recognition from raw audio
  8. How does T5 frame many NLP tasks for one encoder-decoder model?
    • As image classification with a shared label set
    • As speech waveforms with no text tokens
    • As separate model architectures for each task
    • As text-to-text inputs and outputs with task descriptions in the input text