Guide-yakabatana mibvunzo · Zvakaoma Nhanho

Flow Matching Quiz

Explore flow matching’s learned vector fields, conditional paths, ODE sampling and relationship to diffusion-based generation.

Nzira dzekutungamira dzinoenderanaFlow Matching
Mubvunzo 1 ye 8

Ko kuyerera kuenzanirana modhi kunodzidzisei kufembera?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Ko kuyerera kuenzanirana modhi kunodzidzisei kufembera?
    • Munda wekumhanya unofambisa sampuli kubva kune ruzha kuenda kune data
    • A classification label yemufananidzo wega wega
    • Iyo yakamanikidzwa JPEG bytes yemufananidzo
    • Tafura yekutarisa yakagadziriswa yemapixel values
  2. Masampuli anogadzirwa sei kubva kune yakadzidziswa kuyerera inoenderana modhi?
    • Nekuenzanisira discrete Markov ketani yemashoko
    • Nekubatanidza iyo yakadzidza velocity ndima neODE solver kutanga kubva kune ruzha
    • Nekumhanyisa gradient kudzika pamufananidzo wekupinza
    • Nekuenzanisa iyo yese yekudzidziswa seti
  3. Chii chinoita kuti mamiriro ekuyerera anoenderana nekurasikirwa akwanise kudzidzisa?
    • Iyo inofuratira iyo data zvachose
    • Zvinoda kugadzirisa iyo yakazara marginal ndima yekuongorora
    • Iyo inomira pane imwechete data point uye inoshandisa iri nyore inozivikanwa nzira yekududzira
    • Inongoshanda pamufananidzo mumwe chete panguva nekusingaperi
  4. Sei iyo 'yakagadziridzwa kuyerera' musiyano ichigonesa kukurumidza kugadzirwa?
    • Inoshandisa yakakura neural network
    • Inodarika kudzidzira zvachose
    • Inoshandura mifananidzo kuita zvinyorwa kutanga
    • Iyo inotwasanudza nzira dzinogona kuitika saka mashoma ODE solver matanho anodiwa
  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. Mune inomisikidzwa yekuyerera inofananidzwa, chii chakajairika kududzira pakati pesampu yeruzha uye data data panguva t?
    • A random shuffle of pixels
    • x_t = (1-t)* ruzha + t*data, musanganiswa wakatwasuka
    • Squaring yega pixel kukosha
    • Kuronga pixels nekupenya
  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