Kedis ku xidhan hagaha · Dhexdhexaad Heerka

Kediska Fikirka badan ee Fikirka ah ee Faafinta Gaarka ah

Dib-u-eegisga fiiro gaar ah u-habboon ee Faafinta Custom, shakhsiyaynta dhawr-shot, habayn, iyo siyaalo la taageeray si loo curiyo fikrado badan.

Wadooyinka haga ee la xidhiidhaHabaynta Fikradaha Kala Duwan ee Gaarka ah
Su'aal 1 ee 8

Qaybtee ka mid ah moodeelka fidinta ayaa faafinta Custom-ka ugu horrayn si fiican u hagaajinayaa?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Qaybtee ka mid ah moodeelka fidinta ayaa faafinta Custom-ka ugu horrayn si fiican u hagaajinayaa?
    • Saadaasha furaha iyo qiimaha ee lakabyada feejignaanta
    • Miisaan kasta oo ku jira U-Net
    • Furaha VAE kaliya
    • Tokenizer-ka qoraalka
  2. Waa maxay Faafinta Gaarka ah ee ugu caansan marka la barbar dhigo hababka hal-fekerka?
    • Soo saarista muuqaal halkii sawiro
    • Abuuritaanka fikrado badan oo dhawaan laga bartay hal goob
    • U baahnid wax sawiro ah haba yaraatee
    • Ku shaqeeya kaliya CPU-yada
  3. What minimum number of training images did the Custom Diffusion paper report using in some experiments?
    • As few as four images
    • At least 500 images
    • Exactly one image in every experiment
    • The paper specifies no concept images
  4. Waa maxay sababta Faafinta Gaarka ahi u adeegsato jaangooyooyin sawiro joogto ah oo ka socda qaybta ballaadhan ee fikradda?
    • Si loo dedejiyo VAE
    • Si loo kordhiyo xallinta sawirka
    • Si meesha looga saaro calaamadaha biyaha
    • Si looga hortago ku-habboonaanta xad-dhaafka ah iyo luqunta luqadda ee ereyga fikradda
  5. Sidee loo wada isticmaali karaa laba fikradood oo kala-duwan oo gaar ah?
    • Adigoo ku dhex milmay iyada oo loo marayo hagaajin xaddidan oo ku saabsan miisaannada la hagaajiyay
    • Marna la iskuma dari karo
    • Marka la isku celceliyo sawirada gelinta
    • Adiga oo dib u tababaraya qaabka buuxa ee xoqan
  6. What proportion of model weights did the Custom Diffusion paper report storing for its tuned subset?
    • About 3% of the model weights
    • All model weights plus a new model copy
    • No trained parameters at all
    • A fixed 50% of the model weights
  7. Besides merging separately tuned models, how does Custom Diffusion support learning multiple concepts?
    • It requires a separate foundation model for every prompt
    • It replaces all cross-attention with a fixed lookup table
    • It can train for multiple concepts jointly
    • It can only learn one concept per model
  8. In a cross-attention layer, what do the tuned text-side key and value projections help associate?
    • Image pixels with internet search results
    • The diffusion timestep with the model's source code
    • Audio frequencies with a document's page number
    • Prompt concepts with features in the image latent representation