What do Vision Transformers Learn? A Visual Exploration
— AK (@_akhaliq) December 14, 2022
abs: https://t.co/0xJ8UyglHP
github: https://t.co/ftJejEIn43 pic.twitter.com/TPi6YjKZac
What do Vision Transformers Learn? A Visual Exploration
— AK (@_akhaliq) December 14, 2022
abs: https://t.co/0xJ8UyglHP
github: https://t.co/ftJejEIn43 pic.twitter.com/TPi6YjKZac
New work on efficient self-supervised learning: data2vec 2.0 pre-trains vision models 16.4x faster than the most popular existing algorithm.
— Michael Auli (@MichaelAuli) December 13, 2022
Blog: https://t.co/pz6XWKOGGh
Paper: https://t.co/x4kgTbMj7t
Code/models: https://t.co/DoUxCzkHEX
with @ZloiAlexei @mhnt1580 @arunbabu1234 pic.twitter.com/zFHnpOo4iI
REVEAL: Retrieval-Augmented Visual-Language Pre-Training with Multi-Source Multimodal Knowledge Memory
— AK (@_akhaliq) December 13, 2022
abs: https://t.co/8e5JXjSZlS pic.twitter.com/Wp0HKjNOCg
CLIP Itself is a Strong Fine-tuner: Achieving 85.7% and 88.0% Top-1 Accuracy with ViT-B and ViT-L on ImageNet
— AK (@_akhaliq) December 13, 2022
abs: https://t.co/T78PJg7rHK pic.twitter.com/SBL4DksY9z
MAGVIT: Masked Generative Video Transformer
— AK (@_akhaliq) December 13, 2022
abs: https://t.co/LOuF71lgLl
project page: https://t.co/o1uf6BsCbB pic.twitter.com/rHfhtUqeBB
Seeing a Rose in Five Thousand Ways
— AK (@_akhaliq) December 12, 2022
abs: https://t.co/g3DPhg4FLY pic.twitter.com/GUZWdmjmQF
Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis
— AK (@_akhaliq) December 12, 2022
abs: https://t.co/nW5Gpx56ov
github: https://t.co/yXx0RJyzhH pic.twitter.com/qXisikcrDO
LORA - Low-rank Adaptation for Fast Text-to-Image Diffusion Fine-tuning @Gradio demo on @huggingface spaces by @yvrjsharma
— AK (@_akhaliq) December 10, 2022
⁰Fine-tune Stable diffusion models twice as faster than dreambooth method, by Low-rank Adaptation
demo: https://t.co/aV5QusLq5s
Just read through state of AI report by McKinsey: https://t.co/fYhnZ2JGvX
— Sebastian Raschka (@rasbt) December 10, 2022
(As a researcher, it seems to be useful summary of how AI is *actually* used in industry.)
Interesting insights
1. Computer vision now ties with NLP for classification/understanding
1 of 4 pic.twitter.com/8FBhjjRJau
VideoDex: Learning Dexterity from Internet Videos
— AK (@_akhaliq) December 9, 2022
abs: https://t.co/j9g9QJ8X2L
project page: https://t.co/2CwgvxlDqo pic.twitter.com/5FunjX44NY
Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Models
— AK (@_akhaliq) December 9, 2022
abs: https://t.co/VfVDCq1cwY pic.twitter.com/IDDDUuhs2b
SINE: SINgle Image Editing with Text-to-Image Diffusion Models
— AK (@_akhaliq) December 9, 2022
abs: https://t.co/t9DOAdJUgU
project page: https://t.co/G4PGZCKeGQ pic.twitter.com/H3ugxkPaR5