[ENLACE AL PAPER]
"Mastering Diverse Domains through World Models" https://
arxiv.org/pdf/2301.04104
v1.pdf
…
MULTIMODAL AI
-
World Models Master Diverse AI Domains
By
–
-
DreamerV3: General AI System Masters Diverse Simulated Environments
By
–
Y ojo, el sistema DreamerV3 tiene mucha más chicha más allá de Minecraft. De hecho, lo relevante es que es lo suficientemente general como para aprender a dominar diversos problemas (entornos simulados en este caso) que le echen. Os dejo enlace al paper en el siguiente tweet.
-
Hugging Face PickaPic Dataset for Image Selection Released
By
–
Very cool to see the linked Hugging Face dataset – https://
huggingface.co/datasets/yuval
kirstain/PickaPic
… Looking forward to checking out the upcoming photo dump this weekend! -
AI Detection Poll Results: Twitter and LinkedIn Comparison
By
–
Yesterday, I made a poll on both twitter linkedin: which of the following images is AI generated. Twitter poll received 1765 votes: 51.2% A, 48.8% B. Linkedin poll received: 1903 votes: 54% A, 46% B. The correct answer is B! Source of A: https://
site.uit.no/ssworkshop/201
9/07/03/visiting-harstad-and-tromso/
…. More such polls? -
Building a Better AI Assistant: Whisper, ChatGPT, VALL-E Stack
By
–
The perfect recipe to build a Siri or Alexa 10x better! 1. Whisper will listen to what you say.
2. ChatGPT will create a response.
3. VALL-E will read the response with the natural voice you want in seconds! We're close to JARVIS! -
Mid-U Guidance: Fast Classifier Guidance for Latent Diffusion
By
–
Check out this powerful new diffusion inference technique from @johnowhitaker – 'Mid-U Guidance': Fast Classifier Guidance for Latent Diffusion Models. It lets you very flexible push your generated images towards almost any concept you can imagine.
-
Google’s Masked Transformers Achieve SOTA Text-To-Image Performance
By
–
Google’s Masked Generative Transformers Achieve SOTA Text-To-Image Performance With Improved Efficiency https://
syncedreview.com/2023/01/09/goo
gles-masked-generative-transformers-achieve-sota-text-to-image-performance-with-improved-efficiency/
… -
Partial Solution to AI Image Attribute Filtering Limitations
By
–
#48 SOLUCIÓN A MEDIAS. Un arreglo parcial al problema del filtrado de atributos es intentar forzar la estética describiendo minuciosamente cada cosa. "Quiero un cactus AZUL en una bota MARRÓN…" Éxito parcial. Mirad incluso cómo la bota incluye características del cactus
-
Attribute Filtering in Text Generation AI Models
By
–
#47 FILTRACIÓN Al igual que el agua que se cuela por las grietas de tu casa, con las IAs de generación por texto existe el concepto de FILTRADO DE ATRIBUTOS. Fíjate en la imagen anterior cómo a pesar de haberle pedirle un CACTUS AZUL, el color se ha filtrado hacia el zapato.
-
Midjourney Fashion Design: Generate Stunning Model Dresses AI
By
–
#44 DISEÑO DE MODA Y MJ se atreve con todo. Puedes probar con conceptos como «Tetris» o «Arrecife de Coral» y sorprendeos con los resultados! ELIGE ESTILO "top model wearing a fashion dress inspired on [ESTILO], studio light, photorealistic, hyperrealistic, 8K"