Ikibazo gihuza ibibazo · Hagati Urwego

Encoder-Decoder Ubwubatsi Ikibazo

Ongera usuzume kodegisi-decoder yicyitegererezo, inkomoko-ku-ntego yo kwitaho hamwe ninyandiko-ku-mwandiko ikoreshwa mubusemuzi n'imirimo y'ururimi.

Inzira zijyanye no kuyoboraEncoder Decoder Ubwubatsi
Ikibazo 1 Bya 8

Nuwuhe murimo wingenzi wa encoder muburyo bwa encoder-decoder?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Nuwuhe murimo wingenzi wa encoder muburyo bwa encoder-decoder?
    • Gusoma ibyinjijwe no kubihindura bikungahaye imbere
    • Kubyara ibisohoka ikimenyetso kimwe icyarimwe
    • Gutondekanya ibisubizo byabakandida kubishoboka
    • Kubika icyitegererezo cyamahugurwa yamakuru
  2. Kuki decoder ikoresha maskike (impamvu) yo kwiyitaho?
    • Guhisha ibyinjijwe muri kodegisi
    • Kurinda buri mwanya kubona ibimenyetso bizaza mugihe cyibisekuru
    • Gucomeka ibisohoka mubimenyetso bike
    • Kugirango uhitemo ibimenyetso kugirango ubisanzwe
  3. Ni ubuhe buryo buhuza decoder na kodegisi ihagarariwe?
    • Kureka
    • Ibisanzwe
    • Kwambukiranya ibitekerezo
    • Kode y'imyanya
  4. Nubuhe muri ubu buryo bwububiko bwa kodegisi-decoder (ikurikirana-ikurikiranye)?
    • GPT-3
    • BERT
    • Ijambo2Vec
    • T5
  5. Ni ukubera iki gushushanya kodegisi-decoder isanzwe ikwiye guhindurwa?
    • Kuberako ibyinjira nibisohoka bitandukanye, bisaba gusobanukirwa byuzuye noneho ibisekuru bishya
    • Kuberako ibyinjira nibisohoka burigihe burebure
    • Kuberako itigera ikenera kubyara inyandiko nshya
    • Kuberako irinda gukoresha ibitekerezo rwose
  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