Ikibazo gihuza ibibazo · Hagati Urwego

GLIDE Diffusion Model Ikibazo

Shakisha GLIDE inyandiko-iyobora ibisekuru hamwe no gushushanya, amahugurwa asabwa, ibyiciro bitarondoreka, hamwe namateka yububiko.

Inzira zijyanye no kuyoboraImodèle ya Glide Diffusion
Ikibazo 1 Bya 8

Nubuhe buryo bwo kuyobora GLIDE yasanze yabyaye amashusho meza, yihuse-yizerwa?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Nubuhe buryo bwo kuyobora GLIDE yasanze yabyaye amashusho meza, yihuse-yizerwa?
    • Ubuyobozi butarangwamo ubuyobozi
    • Ubuyobozi bwa CLIP
    • Nta buyobozi
    • Guhindura intoki
  2. What size did the published full GLIDE base diffusion model have, before counting its upsampler?
    • About 3.5 thousand parameters
    • About 350 billion parameters
    • About 3.5 billion parameters
    • No trainable parameters
  3. Nigute ubuyobozi butagira ibyiciro bukora mugihe cy'amahugurwa?
    • Icyitegererezo cyigera kibona ikibazo gikwiye
    • Ishusho yihariye itondekanya kuruhande
    • Ibisobanuro bihindurwa kumibare uko bishakiye
    • Icyitegererezo rimwe na rimwe kibona ikibazo nyacyo rimwe na rimwe kikaba ari ubusa
  4. Which later system describes its diffusion decoder as modifying the GLIDE architecture in the DALL-E 2 research paper?
    • The original text-only GPT-2 model
    • The unCLIP system associated with DALL-E 2
    • The Whisper speech recognizer
    • The CLIP contrastive encoder alone
  5. Kuki OpenAI yasohoye gusa verisiyo ntoya, yungurujwe ya GLIDE kumugaragaro?
    • Icyitegererezo cyuzuye cyatakaye
    • Yakoraga kuri mudasobwa zidasanzwe
    • Impungenge zijyanye no gukoresha nabi amashanyarazi akomeye
    • Uruhushya rutangwa nundi muntu
  6. Ni ubuhe buryo bufatika bwo guhindura GLIDE yashyigikiye?
    • Gushushanya inyandiko
    • Kwandika amajwi
    • Urupapuro rwerekana urupapuro
    • Gutanga amashusho
  7. How did GLIDE's upsampler complement its base text-conditioned diffusion model?
    • It replaced the text prompt with an audio transcript
    • It increased the resolution of the image produced by the base model
    • It converted the image into the training tokenizer's vocabulary
    • It supplied manual labels for every generated object
  8. At GLIDE sampling time, how does classifier-free guidance strengthen prompt conditioning?
    • It retrains the model weights after every generated pixel
    • It requires a separate class-label classifier to score every step
    • It replaces the conditional prediction with the unconditional one
    • It combines conditional and unconditional predictions, extrapolating toward the conditional one