Ikibazo gihuza ibibazo · Hagati Urwego

Gutanga inyandiko mu mashusho yakozwe na AI Ikibazo

Gerageza impamvu CLIP subword tokens yahishe imyandikire, uburyo T5 na byte-urwego rwa kodegisi yazamuye inyuguti, nuburyo amahugurwa ya glyph azi neza no gushushanya.

Ikibazo 1 Bya 8

Kuki amamoderi y'amashusho ashingiye kuri CLIP yagize ikibazo cyo kwandika amagambo neza?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Kuki amamoderi y'amashusho ashingiye kuri CLIP yagize ikibazo cyo kwandika amagambo neza?
    • CLIP had no vocabulary at all
    • CLIP turns prompts into subword tokens that capture meaning, not letter sequences
    • CLIP processes only numbers
    • CLIP was trained only on images that contained no text
  2. Ni ubuhe bwoko bw'umusobanuzi ubushakashatsi bushingiye ku nyuguti bwa Google bwasanze bwandika neza?
    • A smaller CLIP encoder
    • An audio encoder
    • An image-only encoder
    • A byte-level encoder such as ByT5 that sees individual characters
  3. Kuki ikosa rito ryihanganirwa mu maso yakozwe ariko ntiryihanganirwe mu nyuguti zakozwe?
    • A single wrong stroke can change which letter it is
    • Faces are generated by a separate model
    • Text is always drawn at a higher resolution
    • Faces use a different kind of noise
  4. Ni iyihe mbogamizi mu makuru y'imyitozo yatumye bigora guhuza amagambo n'imiterere y'inyuguti?
    • Training images used too many fonts
    • All training images were black and white
    • Captions rarely wrote out the text visible in the image
    • Only English-language images were included
  5. Ni iki sisitemu z'ubushakashatsi nka TextDiffuser na GlyphControl zongereye mu gukora amashusho?
    • Invisible watermarks
    • Explicit layouts or pre-rendered letter images to guide the model
    • Audio narration
    • Faster sampling and nothing else
  6. Which model does the guide say paired its image generator with a T5 encoder?
    • Stable Diffusion 3
    • Gutandukana gushikamye 1.5
    • Stable Diffusion 1.4
    • The original CLIP model
  7. For production work that needs exact text, what does the guide call the most reliable approach?
    • Writing a much longer prompt
    • Kugabanya imyanzuro
    • Adding a heavy negative prompt
    • Adding text outside the model, or inpainting a masked area with the corrected word
  8. According to the guide, how do modern image models produce letters?
    • By copying shapes from a font file
    • By running OCR on the prompt
    • By generating them as pixels, which is why long passages and small print still go wrong
    • By looking up images of words on the web