Ikibazo gihuza ibibazo · Biragoye Urwego

Ikibazo cya XLNet

Kugenzura impushya zo kwemeza, ibyerekezo-bibiri byitondewe, Transformer-XL isubiramo, nuburyo ibipimo byerekana biterwa nigenamiterere rigereranijwe.

Inzira zijyanye no kuyoboraIcyitegererezo cya Xlnet
Ikibazo 1 Bya 8

Niyihe ntego yo kwitegura XLNet ikoresha?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Niyihe ntego yo kwitegura XLNet ikoresha?
    • Ururimi rwerekana uruhushya
    • Kwerekana imiterere y'ururimi
    • Gusimbuza ibimenyetso
    • Guhanura interuro ikurikira
  2. Ni izihe ntege nke za BERT XLNet yagenewe gukemura?
    • BERT yari nto cyane
    • Ikimenyetso [MASK] gitera kwitiranya / guhuza neza no gufata ubwigenge bwa maskike-token
    • BERT ntabwo yashoboye gukora interuro ngufi
    • BERT yabuze kodegisi
  3. Ni ubuhe butumwa bwa XLNet bwo-kwiyitaho butandukanye?
    • Encoder na decoder
    • Amahugurwa n'ibishushanyo mbonera
    • Umugezi wibirimo hamwe numurongo wibibazo
    • Gushyiramo interuro ninteruro
  4. Ese XLNet ihindura umubiri mumagambo?
    • Nibyo, biratondekanya ibimenyetso bifatika
    • Nibyo, ariko mugihe cyo gutunganya neza
    • Oya, ntabwo ihindura ibyateganijwe na gato
    • Oya, yemerera gutondekanya ibintu ukoresheje mask yo kwitondera mugihe ukomeje imyanya
  5. Ni ubuhe bwubatsi bukora nkumugongo wa XLNet kumurongo muremure?
    • Guhindura-XL
    • LSTM
    • Umwimerere wa vanilla Transformer
    • Umuyoboro uhuza
  6. How does XLNet combine bidirectional context with autoregressive prediction?
    • It predicts all tokens independently with no attention mask
    • It predicts in sampled factorization orders without inserting a [MASK] input token, using attention masks
    • It physically shuffles the sentence before every prediction
    • It replaces autoregressive prediction with a denoising autoencoder
  7. For a token sequence A B C, suppose a sampled factorization order predicts C before A and B. What context is available when predicting C?
    • The tokens A and B must always be provided first
    • Only the next sentence is available
    • No earlier tokens in that sampled order, although C remains at its original input position
    • The input sequence is physically reordered to C A B
  8. How does XLNet prevent the target token from leaking into its own prediction query?
    • It removes the target from the entire training sequence
    • It uses the target’s label as an input embedding
    • It replaces the token with a [MASK] symbol
    • The query stream represents the target position without its token content