MULTIMODAL AI - AI Dynamics

AI Transforms Filmmaking: Creating Years of Work in Minutes

By

–

15 January 2023 19h30

Foundation #AI goes to the movies
"A.I. could create in minutes what a filmmaker worked on for years" https://
nytimes.com/interactive/20
23/01/13/opinion/jodorowsky-dune-ai-tron.html?searchResultPosition=1
… by @frankpavich

→ View original post on X — @erictopol

15 January 2023

Scaling Laws for Generative Mixed-Modal Language Models

By

@dair_ai

–

15 January 2023 18h31

Introduces scaling laws for generative mixed-modal language models. 9 of 11

→ View original post on X — @dair_ai

15 January 2023

New Multimodal Deep Learning Book Published on ArXiv

By

@dair_ai

–

15 January 2023 18h31

Multimodal deep learning is a new book published on ArXiv. 4 of 11

→ View original post on X — @dair_ai

15 January 2023

Top Machine Learning Papers of the Week

By

@dair_ai

–

15 January 2023 18h31

Top ML Papers of the Week (Jan 9-15): – DreamerV3
– DeepMatcher
– Multimodal deep learning
– Transformer compiler for RASP
– Potential misuses of LMs and mitigations
– Scaling laws for generative mixed-modal LMs
– Time series forecasting with generative modeling
… 1 of 11

→ View original post on X — @dair_ai

15 January 2023

Phenaki text-to-video AI synthesizes realistic videos from prompts

By

@aibreakfast

–

15 January 2023 16h33

📽️Text-to-Video?

It could revolutionize entertainment as we know it.

Here's Phenaki, a model that can synthesize realistic videos from text prompt sequences.

More examples below ↓ pic.twitter.com/kcTHXuQ4Bb
— AI Breakfast (@AiBreakfast) 15 janvier 2023

Text-to-Video? It could revolutionize entertainment as we know it. Here's Phenaki, a model that can synthesize realistic videos from text prompt sequences. More examples below ↓

→ View original post on X — @aibreakfast

15 January 2023

Phenaki compresses video to discrete tokens with causal attention

By

@aibreakfast

–

15 January 2023 16h33

Phenaki, developed by Google Research, compresses the video to a small representation of discrete tokens. This tokenizer uses causal attention in time, which allows it to work with variable-length videos. pic.twitter.com/3mrxF938sM
— AI Breakfast (@AiBreakfast) 15 janvier 2023

Phenaki, developed by Google Research, compresses the video to a small representation of discrete tokens. This tokenizer uses causal attention in time, which allows it to work with variable-length videos.

→ View original post on X — @aibreakfast

15 January 2023

Phenaki joint training on image-text and video-text enables generalization beyond video datasets

By

@aibreakfast

–

15 January 2023 16h33

Phenaki demonstrates how joint training on a large corpus of image-text pairs as well as a smaller number of video-text examples can result in generalization beyond what is available in the video datasets. pic.twitter.com/q100I6PYqa
— AI Breakfast (@AiBreakfast) 15 janvier 2023

Phenaki demonstrates how joint training on a large corpus of image-text pairs as well as a smaller number of video-text examples can result in generalization beyond what is available in the video datasets.

→ View original post on X — @aibreakfast

15 January 2023

DeepFloyd AI’s Text-to-Image Breakthrough Will Transform Graphic Design

By

@swyx

–

15 January 2023 14h19

it is obvious that the @deepfloydai team have made an earth-shattering breakthrough in text-to-image-with-text that will melt the face of visual graphics designers. do not sleep on this. it is going to be as big a leap as going from the shitty GAN era to @StableDiffusion itself.

→ View original post on X — @swyx

15 January 2023