Cuestionario vinculado a una guía · duro Nivel

Prueba de memorización en modelos de difusión

Prueba qué significa la memorización en modelos de difusión, por qué la causan los datos de entrenamiento duplicados y cómo los investigadores extraen y mitigan las casi copias.

Rutas de guía relacionadasMemorización en modelos de difusión
Pregunta 1 de 8

¿Qué es la memorización en un modelo de difusión?

Vista previa de cada pregunta8 preguntas

Las respuestas y explicaciones aparecen a medida que realiza el cuestionario.

  1. ¿Qué es la memorización en un modelo de difusión?
    • The model remembering a user's previous prompts
    • The model reproducing specific training images almost exactly
    • The model storing images in a database for search
    • The model forgetting how to draw after training
  2. ¿Por qué los datos duplicados aumentan la memorización?
    • Duplicates are stored in higher resolution
    • Duplicates are removed automatically before training
    • When an image appears many times, storing it becomes the easiest way to reduce training loss on it
    • Duplicates confuse the model so it outputs noise
  3. ¿Qué demostró el artículo de 2023 de Carlini y sus colegas?
    • Diffusion models never copy training data
    • Only text models memorize data
    • Memorization only happens in black and white images
    • Over a hundred near-identical training images could be extracted from Stable Diffusion
  4. ¿Cómo identificó el ataque de extracción imágenes probablemente memorizadas?
    • It looked for near-identical generations across many seeds for the same caption
    • It read the model's weights directly as images
    • It asked the model which images it had memorized
    • It searched the web for similar photos
  5. ¿Qué afirmación sobre la memorización es correcta?
    • It is the model's normal behavior for every prompt
    • It is rare but can be targeted by a determined user
    • It only occurs as exact pixel copies
    • It has no relevance to privacy
  6. Why is privacy risk sharper when fine-tuning on a small dataset such as medical scans?
    • Small datasets are always public
    • Medical scans are too large to train on
    • Each example has more influence on the model, raising the chance of regurgitation
    • Fine-tuning deletes the base model
  7. Which mitigation bounds the influence of any single training example but currently reduces image quality?
    • Differentially private training
    • Increasing the number of sampling steps
    • Using longer prompts
    • Training for more epochs
  8. Why is deduplication more effective when it removes near-duplicates, not only exact copies?
    • Exact copies do not affect training
    • Near-duplicates are illegal to store
    • Near-duplicates make images blurrier
    • Slightly edited versions of the same image still act as repeated exposure