Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
18
3
15
John Locke
johnlockejrr
Follow
pburub's profile picture
starride-teklia's profile picture
2 followers
Ā·
16 following
[email protected]
AI & ML interests
NLP, OCR, AI
Recent Activity
reacted
to
singhsidhukuldeep
's
post
with š
3 days ago
Exciting breakthrough in AI: @Meta's new Byte Latent Transformer (BLT) revolutionizes language models by eliminating tokenization! The BLT architecture introduces a groundbreaking approach that processes raw bytes instead of tokens, achieving state-of-the-art performance while being more efficient and robust. Here's what makes it special: >> Key Innovations Dynamic Patching: BLT groups bytes into variable-sized patches based on entropy, allocating more compute power where the data is more complex. This results in up to 50% fewer FLOPs during inference compared to traditional token-based models. Three-Component Architecture: ā¢ Lightweight Local Encoder that converts bytes to patch representations ā¢ Powerful Global Latent Transformer that processes patches ā¢ Local Decoder that converts patches back to bytes >> Technical Advantages ā¢ Matches performance of Llama 3 at 8B parameters while being more efficient ā¢ Superior handling of non-English languages and rare character sequences ā¢ Remarkable 99.9% accuracy on spelling tasks ā¢ Better scaling properties than token-based models >> Under the Hood The system uses an entropy model to determine patch boundaries, cross-attention mechanisms for information flow, and hash n-gram embeddings for improved representation. The architecture allows simultaneous scaling of both patch and model size while maintaining fixed inference costs. This is a game-changer for multilingual AI and could reshape how we build future language models. Excited to see how this technology evolves!
liked
a Space
12 days ago
sivan22/Ituria
new
activity
15 days ago
Gabriel/Qwen2-VL-2B-Instruct:
Model inference
View all activity
Organizations
None yet
johnlockejrr
's activity
All
Models
Datasets
Spaces
Papers
Collections
Community
Posts
Upvotes
Likes
liked
a Space
12 days ago
Running
5
š
Ituria
liked
a Space
about 1 month ago
Running
on
T4
6
š
Multicentury HTR Pipeline
Demo for handwritten text recognition model.
liked
a Space
about 2 months ago
Running
4
āļø
PyLaia HTR
liked
a model
2 months ago
MohamedRashad/arabic-small-nougat
Image-to-Text
ā¢
Updated
27 days ago
ā¢
4.29k
ā¢
18
liked
a Space
2 months ago
Running
on
Zero
21
š„
Arabic Auto Tashkeel
liked
a Space
3 months ago
Running
14
š
OPUS Translate
liked
a model
3 months ago
pburub/wav2vec2-xls-r-300m-mwa-maaloula
Automatic Speech Recognition
ā¢
Updated
May 16
ā¢
7
ā¢
1
liked
a Space
3 months ago
Running
249
š
Chat-with-GPT4o-mini
liked
a Space
5 months ago
Running
on
Zero
5.97k
š„ļø
FLUX.1 [dev]
liked
a model
6 months ago
Teklia/doc-ufcn-generic-historical-line
Image Segmentation
ā¢
Updated
Sep 10
ā¢
18
liked
2 Spaces
6 months ago
Sleeping
6
š¢
PyLaia
Running
7
š¢
Doc-UFCN
liked
2 models
8 months ago
dicta-il/dictalm2.0
Text Generation
ā¢
Updated
Jul 10
ā¢
8.47k
ā¢
11
dicta-il/dictalm2.0-GGUF
Text Generation
ā¢
Updated
Jul 10
ā¢
191
ā¢
4
liked
a model
9 months ago
dicta-il/BEREL_2.0
Fill-Mask
ā¢
Updated
Jul 3, 2023
ā¢
31
ā¢
2