Ikibazo gihuza ibibazo · Biroroshye Urwego

Ikibazo cyo Kwerekana Ururimi

Gerageza autoregressive probabilite, n-garama imipaka, intego zikomoka kumibare, ibisekuruza, igihombo, icyitegererezo, indorerezi zamahugurwa, hamwe nibishoboka.

Inzira zijyanye no kuyoboraKwerekana Ururimi
Ikibazo 1 Bya 8

Muri iyo nyandiko y'ikigereranyo cy'ururimi rw'ikora, iki giteganijwe giturutse ku nyuguti z'ibanze?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Muri iyo nyandiko y'ikigereranyo cy'ururimi rw'ikora, iki giteganijwe giturutse ku nyuguti z'ibanze?
    • Igabanyamiterere ry'ubushoboka ku nyuguti ikurikira
    • Urutonde rw'inyuguti mu nyandiko y'umwimerere
    • Igipimo cy'ubushyuhe gusa giturutse ku gikoresho
    • Ikirango cy'ishusho cy'ibanze kitagira amakuru y'urutonde
  2. Ni izihe mbogamizi zigomba kwitabwaho ku rugero ruto rw'inkuru ya n-gram y'ururimi?
    • A requirement that all n-gram models reject every unseen sequence even with smoothing
    • A guarantee of unlimited history
    • A requirement to process only images
    • Limited context and possible zero probabilities for unseen continuations without smoothing
  3. Why can next-token language-model pretraining obtain targets without a separate class annotation for every position?
    • The model has no training objective
    • The target next tokens are already present in the observed text
    • All training text must have been generated by the same model
    • Every position requires an external image classifier
  4. How does autoregressive text generation use a selected continuation token?
    • It predicts every future token independently in one step
    • It discards every selected token before the next prediction
    • It appends the token and uses available earlier output as context for later predictions
    • It changes the training labels to future deployment outcomes
  5. Which objective is commonly minimized for the stated autoregressive next-token training setup?
    • Squared numerical distance between alphabetical token ID values
    • Only the number of tokens in an output filename
    • Only hinge loss on image-pixel locations
    • Cross-entropy or negative log-likelihood for observed target tokens
  6. What do temperature and top-p choices affect during sampled language-model generation?
    • Uko token z'ikurikiranya zihitamo hakurikijwe amanota ya model
    • Umubare w'ibice byigishijwe nyuma ya buri token
    • Garanti ko ibivugwa byose by'output bihinduka ukuri
    • Ikirango cyanditswe muri buri ntego y'amasomo yabonetse
  7. Muri teacher-forced next-token training, konteks y'imbere ya token y'intego iva he?
    • Ni output y'imbere ya model yakuwe mu buryo bw'ubuntu
    • Corpus isimburwa y'ukuri yihinduranya kuri buri mwanya
    • Inyigisho yabonetse
    • Ibimenyetso bya nyuma by'isesengura bitangwa nyuma yo gushyira mu bikorwa
  8. Prefiksi y'ibimenyetso bibiri ivugwa ifite amahirwe ya 0.5 kuri token ya mbere na 0.2 kuri iya kabiri uhereye kuri iya mbere. Ni ikihe kigero cy'ibyago byayo byombi?
    • 0.7
    • 0.1
    • 0.4
    • 0.3