Gids-gekoppelde quiz · Moeilijk Niveau
Glazuur- en nachtschade-quiz voor kunstenaarsbescherming
Test hoe Glaze-stijl cloaking en Nightshade-gegevensvergiftiging gebruik maken van geoptimaliseerde verstoringen, en waarom opschaling en detectoren deze bescherming kunnen wegnemen.
Vraag 1 van 8
Welke dreiging is Glaze in de eerste plaats bedoeld om tegen te gaan?
Bekijk elke vraag8 vragen
Antwoorden en uitleg verschijnen terwijl je de quiz doet.
- Welke dreiging is Glaze in de eerste plaats bedoeld om tegen te gaan?
- Theft of physical paintings
- Style mimicry through fine-tuning a model on an artist's work
- Screenshots of artwork on social media
- Deletion of artwork from websites
- Hoe verschilt Nightshade van Glaze in zijn doel?
- Nightshade watermarks images for copyright registration
- Nightshade encrypts images so nobody can view them
- Nightshade poisons training data so models learn wrong associations between concepts and images
- Nightshade blocks web crawlers from a website
- Wat is het optimalisatiedoel van Glaze wanneer het een kunstwerk verstoort?
- A version of the artwork converted into a different style
- A blank white image
- A random image from the internet
- The artist's signature
- Wat zorgt ervoor dat de verstoringen niet duidelijk zijn voor menselijke kijkers?
- The tools only change image metadata
- The changes are applied only to the image border
- The tools reduce image resolution
- A perceptual constraint, such as an LPIPS budget, limits visible change
- Wat heeft een onderzoek uit 2024 door ETH Zürich en Google DeepMind-onderzoekers gevonden?
- Glaze makes images impossible to train on under any conditions
- Simple methods like noisy upscaling could remove Glaze-style protections enough for mimicry
- Nightshade works only on printed art
- Perturbations become stronger over time
- Why is protection from these tools not permanent?
- A published image cannot be updated while attack methods and models keep changing
- The software licenses expire after a year
- The perturbations fade from the file over time
- Models are required by law to ignore protected images
- What does transferability mean in the context of these tools?
- Moving an image from one website to another
- Transferring copyright to the tool's developers
- Converting an image between file formats
- A perturbation optimized against one encoder still working on other models
- Why did the Nightshade researchers argue a modest number of poisoned images could affect a model?
- Because every model is trained on only 100 images
- Because poisoned images are copied automatically across the internet
- Because each concept appears in relatively few training images compared with the whole dataset
- Because models trust poisoned images more than clean ones