Maswali yanayounganishwa na mwongozo · Kati Kiwango

Maswali ya Mwongozo Isiyo na Kiainishaji

Kagua mwongozo usio na uainishaji, kuacha shule kwa masharti, utofauti wa ubora, gharama za makisio na kiwango cha mwongozo. Angalia upeo na mawazo.

Njia za mwongozo zinazohusianaMwongozo wa Bure wa Kuainisha
Swali 1 ya 8

Why do diffusion models apply classifier-free guidance?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Why do diffusion models apply classifier-free guidance?
    • Ili kuimarisha jinsi matokeo yanavyofuata kwa ukaribu upesi wa uwekaji
    • Ili kukandamiza mfano kwa saizi ndogo
    • Ili kuondoa kisimbaji maandishi kabisa
    • Ili kubadilisha picha kuwa sauti
  2. Mfano hujifunzaje kufanya utabiri usio na masharti kwa mwongozo usio na uainishaji?
    • Kwa kumfundisha mwanafunzi wa darasa la pili
    • Kwa kuacha nasibu hali wakati wa mafunzo
    • Kwa kutumia picha nyeusi tu
    • Kwa kuzima GPU
  3. Je, kuongeza kiwango cha mwongozo kwa ujumla hufanya nini?
    • Daima inaboresha utofauti
    • Huondoa ushawishi wa kidokezo
    • Huimarisha ufuasi wa haraka lakini inaweza kusababisha kujaa kupita kiasi
    • Hufanya picha kuwa ya kijivu
  4. Takriban ni pasi ngapi za mbele kwa kila hatua ya denoising mwongozo wa kawaida usio na uainishaji unahitajika?
    • Sifuri
    • Kumi
    • Mia moja
    • Mbili (moja kwa masharti, moja bila masharti)
  5. Je, mwongozo usio na uainishaji una faida gani juu ya mwongozo wa uainishaji?
    • Haihitaji kiainishi kilichofunzwa tofauti
    • Inahitaji data zaidi ya mafunzo
    • Inafanya kazi tu kwa mifano ya maandishi
    • Huondoa mchakato wa kueneza
  6. In the common classifier-free guidance formula, what does guidance scale s=0 produce?
    • The conditional prediction with extra amplification
    • The unconditional model prediction without conditioning
    • A prediction from a separately trained classifier
    • A sample with all denoising removed
  7. How does classifier-free guidance combine model outputs at inference?
    • It combines two image classifiers
    • It removes the conditional prompt
    • It combines conditional and unconditional score estimates
    • It adds a learned reward model after decoding
  8. What does a guidance scale of 1 generally correspond to in the common formulation?
    • The unconditional prediction only
    • A classifier-based gradient
    • A zero-noise image
    • The conditional prediction without extra guidance amplification