Guide-yakabatana mibvunzo · Pakati Nhanho

Encoder-Decoder Architectures Quiz

Review encoder-decoder model components, source-to-target attention and text-to-text uses in translation and language tasks.

Nzira dzekutungamira dzinoenderanaEncoder Decoder Architectures
Mubvunzo 1 ye 8

Nderipi basa guru reiyo encoder mune encoder-decoder modhi?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Nderipi basa guru reiyo encoder mune encoder-decoder modhi?
    • Kuverenga iyo yekuisa uye kuishandura kuita yakapfuma yemukati inomiririra
    • Kugadzira chinobuda chiratidzo chimwe panguva
    • Kuronga zvabuda nemumiriri nezvinogoneka
    • Kuchengetedza data rekudzidzisa remuenzaniso
  2. Sei decoder ichishandisa masked (causal) kuzvitarisa?
    • Kuviga mapindiro kubva kune encoder
    • Kudzivirira chinzvimbo chimwe nechimwe kubva pakuona zviratidzo zveramangwana panguva yechizvarwa
    • Kudzvanya zvinobuda kuita ma tokens mashoma
    • Kudonhedza zvisina tsarukano ma tokens ekugadzirisa
  3. Ndeipi michina inobatanidza decoder kune zvinomiririra encoder?
    • Dropout
    • Batch normalization
    • Cross-attention
    • Positional encoding
  4. Ndeupi weaya mamodheru ari encoder-decoder (kutevedzana-kune-kutevedzana) kuvaka?
    • GPT-3
    • BERT
    • Word2Vec
    • T5
  5. Sei dhizaini yeencoder-decoder iri yakajairika pakushandura?
    • Nekuti mapindiro nekubuda kwakasiyana kutevedzana, kunoda kunzwisiswa kuzere kozoita chizvarwa chitsva
    • Nekuti kupinza uye kubuda zvinogara zvakafanana kureba
    • Nekuti haimbodi kugadzira zvinyorwa zvitsva
    • Nokuti inodzivisa kushandisa pfungwa zvachose
  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