David Solís retweeted
Pues eso, que he estado probando la IA esta que traduce directamente a cualquier idioma y creo que ya vivimos en 2080
David Solís retweeted
Cuando llegaba la tormenta a Isla Cristina. Fotografías de Antonio G. Alonso (facebook.com/antonio.alonso.…)
David Solís retweeted
Cool work, have a look! Interesting to see they tie the "probability" of discrete representation to, well, the probability of the dataset : Variational Inference itself.
arxiv.org/abs/2304.04820
David Solís retweeted
WTF: Mind reading is here.
Researchers invented a new #AI method to convert brain signals into video. See the results for yourself
Published in Nature yesterday: nature.com/articles/s41586-0…
What are the implications? Is this the biggest paper of 2023?
#CEBRA
David Solís retweeted
Result: quality similar to Copilot at launch!
- Outperform every open code LLM that supports multiple programming languages
- Matches/outperforms Copilot at launch
- 40% pass@1 on HumanEval
David Solís retweeted
Tracking through Containers and Occluders in the Wild
Proposes a model, called TCOW, that can segment objects in videos with a notion of object permanence.
proj: tcow.cs.columbia.edu/
abs: arxiv.org/abs/2305.03052
David Solís retweeted
How to summarize a book without sending 100% of your tokens to a LLM
I tested out a fun method on "Into Thin Air" (~140K tokens)
The results were surprising good
Definitely enough to refine and keep working Here's a speed run though process:
David Solís retweeted
ZipIt! Merging Models from Different Tasks without Training
Presents ZipIt!, a general method for merging two arbitrary models of the same architecture that incorporates two simple strategies.
arxiv.org/abs/2305.03053
David Solís retweeted
Today we are releasing a CLIP ViT-L/14 model with 79.2% zero-shot accuracy on ImageNet.
Our model outperforms OpenAI's CLIP by a large margin, and outperforms even bigger models (ViT-g/14) trained on LAION-2B
Check it out at huggingface.co/laion/CLIP-Vi…!
David Solís retweeted
BREAKING: The future of object detection is here!
Meet YOLO-NAS, open-sourced YOLO-based architecture redefining state-of-the-art in object detection.
David Solís retweeted
Real-Time Radiance Fields for Single-Image Portrait View Synthesis
abs: arxiv.org/abs/2305.02310
project page: research.nvidia.com/labs/nxp…
David Solís retweeted
Leaked Google document: “We Have No Moat, And Neither Does OpenAI”
The most interesting thing I've read recently about LLMs - a purportedly leaked document from a researcher at Google talking about the huge strategic impact open source models are having
simonwillison.net/2023/May/4…
My research is available on @ResearchGate: researchgate.net/publication…


























