Kediska Habboon ee socodka

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Muxuu si toos ah u bartaa qaabka u dhigma socodka si loo saadaaliyo?
   • Goob xawaare ah oo ka soo qaada muunado ka soo qaada qaylada una gudbisa xogta
   • Summada kala soocidda sawir kasta
   • JPEG-bytes-ka la cufan ee sawirka
   • Shaxda raadinta go'an ee qiimaha pixel
2. Sidee shaybaarka looga soo saaraa qaabka u dhigma socodka tababaran?
   • Adiga oo tusaale u soo qaatay silsilad erayo ah oo kala duwan
   • Adigoo ku dhex dara goobta xawaaraha la bartay iyo xaliye ODE oo ka bilaabmaya buuqa
   • Adigoo ku ordaya soo degitaanka gradient ee sawirka gelinta
   • Marka la isku celceliyo dhammaan tababarka
3. Maxaa ka dhigaya luminta qulqulka shuruudaysan ee la jaanqaadi kara in la tababaro?
   • Waxay iska indhatiraysaa xogta gebi ahaanba
   • Waxay u baahan tahay in si buuxda loo xalliyo goobta marginal-ka oo dhan
   • Waxay ku xidhan tahay hal dhibic oo xog ah waxayna isticmaashaa waddo isku xidhidh oo fudud oo la yaqaan
   • Waxa ay ku shaqaysaa hal sawir markiiba waligeed
4. Waa maxay sababta 'qulqulka la hagaajiyay' kala duwanaanshiyaha u suurtageliyaa jiilka degdega ah?
   • Waxay isticmaashaa shabakadaha neerfaha ee waaweyn
   • Waxa ay gabi ahaanba ka boodo tababarka
   • Marka hore waxay u rogtaa sawirada qoraal
   • Waxay toosisaa dariiqyada ixtimaalka si ay yar tahay tillaabooyin xaliya ODE ayaa loo baahan yahay
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. Marka la eego qulqulka shuruudaha shuruudaysan, waa maxay isdhexgalka caadiga ah ee u dhexeeya muunadda sanqadha iyo muunadda xogta wakhtiga t?
   • Isku shaandheyn random ah oo pixels ah
   • x_t = (1-t)* buuq + t*xog, isku darka xariiq toosan
   • Isku dhufashada qiimaha pixel kasta
   • Ku kala soocida pixels iftiin
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