quiz bu lëkkale ak tegtal · Dëgër Niveau

Quiz buy méngoo

Saytu barabi vecteur yiñ jàngee ci méngoo, yooni sart, echantillonnage ODE ak lëkkaloo gi am ci diffusion-based generation.

Yooni njiit yu jëm ciméngoo
Laaj 1 de 8

Lan la modelu matching flow di jàng ngir wax luy waaja am?

Xoolaatal laaj bu nekk8 laaj

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

  1. Lan la modelu matching flow di jàng ngir wax luy waaja am?
    • Benn gaawaay buy toxal misaal yi ci bruit bi dem ci done yi
    • Etiketu xaaj ngir nataal bu nekk
    • Byte JPEG yiñ komprime ci benn nataal
    • Tablo seetlu bu takku bu valeur pixel
  2. naka lañuy defaree misaal ci xeetu matching flow buñ tàggat?
    • Ni ñuy jëlee misaalu benn chaîne de mots Markov
    • Ni ñu boole gaawaay bi ñu jàng ak ODE solver bu tàmbalee ci bruit
    • Soo defee wàccinu gradient ci nataalu dugal bi
    • Sooy jël moyenne bi ci tàggat yaram bi yépp
  3. Lan moo waral debit bi méngoo ak perte bi mën nañu ko tàggat?
    • Dafay baña bayyi xel ci done yi
    • Dafay laaj nga saafara barabu marginal bi yépp ci anam wu analitik
    • Dafay tënku ci benn poñ de done, ba noppi jëfandikoo yoonu interpolaasioŋ bu yomb
    • Benn nataal kese lay dox saa yu nekk ba fàww
  4. Lan moo waral 'flow rectified' bi gëna gaaw ci defar?
    • Dafay jëfandikoo reso neuronal yu gëna mag
    • Dafay sànni tàggat yaram
    • Dafay njëkka soppi nataal yi ci mbind
    • Dafay jubbanti yooni probabilite yi suko defee ñu gëna néew luñu soxla ci ODE solver
  5. Which named image model’s paper explicitly describes a rectified-flow transformer?
    • Stable Diffusion 3
    • Flux, as established by the cited source set
    • Resnet-50
    • YOLOv8
  6. Ci méngoo debit bu am sart, lan mooy interpolation bi gëna mag ci digganté misaalu bruit ak misaalu done ci waxtu t?
    • njaxas pixel yu bari
    • x_t = (1-t)*bruit + t*done, njaxasu liiñ bu jub
    • Kaare valeur pixel bu nekk
    • Tànn pixel yi ci leeraay bi
  7. In conditional flow matching, how is the marginal vector field related to conditional vector fields?
    • It is the largest vector in the training batch
    • It is a fixed class label independent of time
    • It is the conditional expectation of the conditional vector field over the conditioning variable
    • It is the average pixel value of the data
  8. Beyond generative modeling, what other task does the original rectified-flow paper describe?
    • Speech diarization with speaker embeddings
    • Classifying handwritten digits from labels alone
    • Predicting calendar availability
    • Data or domain transfer between distributions