Rita Wu retweetledi
Rita Wu
97 posts

Rita Wu
@digwu_
CRUNCHY, DIONYSIAC, EXPERIMENTAL, ENTROPIC, "INDISCIPLINARY", DISRUPTIVE AND CAFFEINATED // Making connections to make a difference
Katılım Ekim 2019
757 Takip Edilen1.2K Takipçiler
Rita Wu retweetledi

Given all the recent buzz around DeepSeek, I decided to re-create a shallow and low-dimensional version of V3's forward pass as a spreadsheet, matching the exact computation and fitting parameters in PyTorch with a synthetic dataset. More details here: linkedin.com/pulse/shallows…
English
Rita Wu retweetledi

o1 is a different kind of model. great performance requires using it in a new way relative to standard chat models.
Dan McAteer@daniel_mac8
this is an amazing way to think about prompting o1 from @benhylak
English
Rita Wu retweetledi
Rita Wu retweetledi

The dog behind the #Doge meme and #Dogecoin cryptocurrency is named Kabosu and she still lives with her owner Atsuko Sato @kabosumama in Sakura, Japan. Kabosu was a rescue dog and became a meme after Atsuko uploaded photos of Kabosu, including the one below, on her blog in 2010.

English
Rita Wu retweetledi
Rita Wu retweetledi

@EnelClientesBR Quero ver depois vir cobrar conta alguma...tinham que pagar multa absurda , indenizar todos os clientes e perder a concessão
Português

@EnelClientesBR Absurdo
Ainda nada...aqui com cachorro é idosa passando mal...nem água gelada tem
Português
Rita Wu retweetledi
Rita Wu retweetledi
Rita Wu retweetledi

Sora can also generate stories involving a sequence of events, although it's far from perfect.
For this video, I asked that a golden retriever and samoyed should walk through NYC, then a taxi should stop to let the dogs pass a crosswalk, then they should walk past a pretzel and hot dog stand, and finally they should end up looking at Broadway signs.
English
Rita Wu retweetledi
Rita Wu retweetledi

Google presents LUMIERE
A Space-Time Diffusion Model for Video Generation
paper page: huggingface.co/papers/2401.12…
Demonstrate state-of-the-art text-to-video generation results, and show that our design easily facilitates a wide range of content creation tasks and video editing applications, including image-to-video, video inpainting, and stylized generation.
English
Rita Wu retweetledi

180 degrees disagree. At best the board at OpenAI had a Pyrrhic victory. Most likely outcome now seems to be Sam and team with fewer constraints under Microsoft. Elite opinion strongly with Sam. Raises questions about their tactical intelligence.
nytimes.com/2023/11/20/tec…

English
Rita Wu retweetledi
Rita Wu retweetledi
Rita Wu retweetledi
Rita Wu retweetledi



















