244
Jupyter Agent 2
๐
Run code and analyze data in a Jupyter notebook
John Johnson
jjokah
elonreevemusk009's profile picture
DannisWrites's profile picture
sometimesanotion's profile picture
ยท
AI & ML interests
Natural Language Processing, AI Engineering, Small Language Models.
Recent Activity
liked
a Space
14 days ago
lvwerra/jupyter-agent-2
liked
a model
28 days ago
ibm-granite/granite-docling-258M
liked
a Space
28 days ago
ibm-granite/granite-docling-258m-demo
Organizations
liked
a
Space
14 days ago
liked
a
model
28 days ago
liked
a
Space
28 days ago
liked
a
Space
about 1 month ago
upvoted
3
changelogs
about 2 months ago
Changelog
Introducing a better Hugging Face CLI
โข
93
Changelog
Trending Papers
โข
104
Changelog
Introducing HF Jobs: Run scalable compute jobs on Hugging Face
โข
200
reacted to
their
post with ๐ฅ
about 2 months ago
Post
322
The combination of Gemini Nano (Google's AI model) and the Tensor G5 chip (Google's AI processor), built into the Pixel 10 (Google's Smartphone), provides Google with a strong foundation to continue pushing the limits of edge AI โ ๐ฎMagic Cue.
Magic Cue digs through your device (Gmail, Calendar, Messages, Photos, screenshots, notes, and more) to surface whatโs useful at that moment.
Ref (Magic Cue):
https://store.google.com/intl/en/ideas/articles/magic-cue/
Magic Cue digs through your device (Gmail, Calendar, Messages, Photos, screenshots, notes, and more) to surface whatโs useful at that moment.
Ref (Magic Cue):
https://store.google.com/intl/en/ideas/articles/magic-cue/
posted
an
update
about 2 months ago
Post
322
The combination of Gemini Nano (Google's AI model) and the Tensor G5 chip (Google's AI processor), built into the Pixel 10 (Google's Smartphone), provides Google with a strong foundation to continue pushing the limits of edge AI โ ๐ฎMagic Cue.
Magic Cue digs through your device (Gmail, Calendar, Messages, Photos, screenshots, notes, and more) to surface whatโs useful at that moment.
Ref (Magic Cue):
https://store.google.com/intl/en/ideas/articles/magic-cue/
Magic Cue digs through your device (Gmail, Calendar, Messages, Photos, screenshots, notes, and more) to surface whatโs useful at that moment.
Ref (Magic Cue):
https://store.google.com/intl/en/ideas/articles/magic-cue/
liked
a
model
2 months ago
commented on
Small Language Models (SLM): A Comprehensive Overview
2 months ago
That's right. XLM-RoBERTa is a beast for the tasks it was optimized for.
posted
an
update
2 months ago
Post
451
After responding to this question on my blog about SLMs, I'm beginning to wonder if the term "Small Language Model" is already dated.
Ref (article):
https://huggingface.co/blog/jjokah/small-language-model
Ref (article):
https://huggingface.co/blog/jjokah/small-language-model
commented on
Small Language Models (SLM): A Comprehensive Overview
2 months ago
Based on size, XML-RoBERTa (which is under 1 billion parameters) might be considered a Small Language Model. However, in comparison to the modern SLMs (w/ broad generative capabilities), XML-RoBERTa is highly specialized for cross-lingual transfer tasks. In that case, it cannot be considered a typical SLM.
upvoted
an
article
2 months ago
Article
Build an AI Shopping Assistant with Gradio MCP Servers
โข
58
liked
a
Space
5 months ago
liked
2
models
6 months ago
upvoted
a
paper
7 months ago
reacted to
their
post with ๐ฅ๐
7 months ago
Post
2410
# Video Tokenization โ for efficient AI video processing
Meet ๐๐ข๐๐๐จ๐ค, a new open-source video tokenization technique developed by Microsoft Research to address the computational challenges of processing large volumes of video data. The core problem VidTok tackles is the inefficiency caused by redundant information in raw video pixels.
VidTok converts complex video footage into compact, structured units called tokens, making it easier and more efficient for AI systems to analyze, understand, and generate video content.
Research Paper: https://arxiv.org/abs/2412.13061
VidTok Code: https://github.com/microsoft/VidTok
Meet ๐๐ข๐๐๐จ๐ค, a new open-source video tokenization technique developed by Microsoft Research to address the computational challenges of processing large volumes of video data. The core problem VidTok tackles is the inefficiency caused by redundant information in raw video pixels.
VidTok converts complex video footage into compact, structured units called tokens, making it easier and more efficient for AI systems to analyze, understand, and generate video content.
Research Paper: https://arxiv.org/abs/2412.13061
VidTok Code: https://github.com/microsoft/VidTok