Ikibazo gihuza ibibazo · Hagati Urwego

Tekereza Inyandiko-Kuri-Ikibazo

Ongera usubiremo umwimerere Imagen itondekanya inyandiko, ibisekuruza byashizwe hejuru, imbaraga zingana, hamwe no guhitamo imbaraga mumahugurwa ya super-resolution.

Inzira zijyanye no kuyoboraTekereza Umwandiko Kuri Ishusho
Ikibazo 1 Bya 8

In the original Imagen study's scaling comparison, which change produced larger image-quality and text-alignment gains?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. In the original Imagen study's scaling comparison, which change produced larger image-quality and text-alignment gains?
    • Discarding the language encoder entirely
    • Increasing the pretrained text encoder's size rather than the diffusion model's size
    • Replacing diffusion with nearest-neighbor image lookup
    • Reducing every prompt to a single word
  2. Niyihe kodegisi ya Imagen yishingikiriza?
    • Urushundura
    • Gitoya LSTM
    • Amategeko yanditse mu ntoki
    • Icyitegererezo cyururimi T5-XXL
  3. Nigute Imagen igera kumurongo wo hejuru?
    • Gukwirakwiza gukwirakwiza hamwe na super-reaction yo kuzamura ibyiciro
    • Intambwe imwe nini yintambwe
    • Mugukata ifoto nini
    • Muguhuza amashusho abiri
  4. What problem did original Imagen's dynamic thresholding address under strong guidance?
    • A lack of text tokens in the encoder vocabulary
    • The need to compress the final PNG file
    • Saturation of predicted pixel values during sampling
    • A missing training label for every object
  5. DrawBench ni iki?
    • Igishushanyo
    • Ikirangantego cya GPU
    • Dataset yamafoto yinjangwe
    • Igipimo cyibibazo bitoroshye byo gusuzuma inyandiko-ku-shusho
  6. Kuki kodegisi ya T5-XXL ikomeza gukonjeshwa muri Imagen?
    • Yavunitse kandi ntishobora kwitoza
    • Inyandiko zayo zikungahaye zisobanutse kuva mbere yo kwimura nta yandi mananiza
    • Kubika umwanya wa disiki gusa
    • Kuberako inyandiko ntaho ihuriye
  7. What resolution did original Imagen's base diffusion model generate before its super-resolution stages?
    • 256 by 256 pixels as the first stage
    • 1024 by 1024 pixels directly
    • 64 by 64 pixels
    • A single pixel that was copied across the final image
  8. Why did the Imagen researchers add Gaussian noise to low-resolution conditioning images during super-resolution training?
    • To make the upsamplers more robust to imperfect inputs from earlier stages
    • To remove text conditioning from the entire cascade
    • To guarantee that all training images had the same caption
    • To replace the upsamplers with deterministic resizing