Ikibazo gihuza ibibazo · Hagati Urwego
Kubogama na Stereotypes mubibazo bya AI Ishusho Yabajijwe
Ikizamini aho amashusho-atera kubogama aturuka, uburyo abashakashatsi babipima nibitekerezo bitabogamye, nimpamvu gukosorwa nko kwandika byihuse bishobora gukosorwa.
Ikibazo 1 Bya 8
Ni ukubera iki ikibazo kitavuze uburinganire cyangwa ubwoko bikibyara ishusho ibogamye?
Reba buri kibazo8 ibibazo
Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.
- Ni ukubera iki ikibazo kitavuze uburinganire cyangwa ubwoko bikibyara ishusho ibogamye?
- The model randomly assigns attributes with equal probability
- The model falls back on the most typical picture it learned for the word, which reflects skewed data
- Users secretly add hidden terms
- Image generators cannot draw people accurately
- Ni ubuhe bwoko bw'amakuru moderi nka Stable Diffusion yigira?
- Hand-drawn images commissioned for training
- Government census photographs
- Large collections of image-caption pairs scraped from the web
- Only stock photos with verified demographics
- Nubuhe buryo busanzwe abashakashatsi bakoresha mugupima kubogama mumashanyarazi?
- Asking the model to rate its own fairness
- Counting how many words are in each prompt
- Checking image file sizes
- Generating many images for neutral prompts and comparing who appears
- Ni ukubera iki umushinga wa Stable Bias wakoresheje cluster aho gushyiramo ubwoko butaziguye?
- Labeling someone's race from an image is itself unreliable and fraught
- Clustering is required by law
- Race labels made the images too large
- The images contained no people
- Isesengura rya 2023 Bloomberg ryakozwe na Stable Diffusion ryabonye iki?
- Outputs were perfectly balanced across all jobs
- Only landscapes showed bias
- High-paying jobs skewed toward lighter-skinned men
- The model refused to draw people at work
- What happened with Google's Gemini image generation in February 2024?
- It was paused after producing historically inaccurate diverse depictions
- It was praised for perfect historical accuracy
- It was banned in all countries
- It stopped generating landscapes
- What is the main weakness of simple behind-the-scenes prompt rewriting as a fix?
- It is too expensive to run
- It can be context-blind and alter prompts that specify a period or identity
- It makes images black and white
- It deletes the training data
- Which statement about the source of bias is accurate according to the guide?
- Bias comes only from training data
- Bias comes only from users' prompts
- Bias is fully solved by larger models
- Data, filtering, captioning, model design and post-training choices all shape bias