Mohamed Rashad PRO
MohamedRashad
AI & ML interests
Computer Vision, Robotics, Natural Language Processing
Recent Activity
liked
a Space
2 days ago
Remsky/Kokoro-TTS-Zero
reacted
to
their
post
with 🚀
7 days ago
The winners of Best Paper Award in NeurIPs2024 (FoundationVision) https://huggingface.co/papers/2404.02905 has just released a new paper called infinty:
https://huggingface.co/papers/2412.04431
And i managed to build a space for it so anyone can try it out: https://huggingface.co/spaces/MohamedRashad/Infinity
The idea of a text to image model using autoregressive archticture is quite interesting in my opinion.
reacted
to
alielfilali01's
post
with 👍
7 days ago
3C3H AraGen Leaderboard welcomes today https://huggingface.co/deepseek-ai/DeepSeek-V3 and 12 other models (including the late gpt-3.5 💀) to the ranking of best LLMs in Arabic !
Observations:
- DeepSeek-v3 ranked 3rd and only Open model among the top 5 !
- A 14B open model (https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) outperforms gpt-3.5-turbo-0125 (from last year). This shows how much we came in advancing and supporting Arabic presence within the LLM ecosystem !
- Contrary to what observed in likelihood-acc leaderboards (like https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard) further finetuned models like https://huggingface.co/maldv/Qwentile2.5-32B-Instruct actually decreased the performance compared to the original model https://huggingface.co/Qwen/Qwen2.5-32B-Instruct.
It's worth to note that the decrease is statiscally insignificant which imply that at best, the out-domain finetuning do not really hurts the model original capabilities acquired during pretraining.
Previous work addressed this (finetuning VS pretraining) but more investigation in this regard is required (any PhDs here ? This could be your question ...)
Check out the latest rankings: https://huggingface.co/spaces/inceptionai/AraGen-Leaderboard
Organizations
MohamedRashad's activity
Update README.md
#1 opened 27 days ago
by
MohamedRashad
ideas about automatic summarization of qur'an-tafseer
3
#2 opened 27 days ago
by
rhyssh
How my country is not in the global top 500 model creators on the Hub ?
#6 opened 30 days ago
by
MohamedRashad
RuntimeError: Unexpected error from cudaGetDeviceCount()
5
#138 opened about 1 month ago
by
bluestyle97
Will you open source the comparison dataset ?
1
#68 opened about 1 month ago
by
MohamedRashad
Arabic Small Nougat
10
#1 opened 9 months ago
by
johnlockejrr
ValueError: Invalid LoRA checkpoint.
2
#5 opened about 2 months ago
by
MohamedRashad
The leaderboard is not working anymore...
1
#2 opened about 2 months ago
by
zhiminy
Create story
8
#1 opened 2 months ago
by
notmebug
Apply for community grant: Personal project (gpu)
#1 opened 11 months ago
by
MohamedRashad
مشكلة التطويل
3
#2 opened 11 months ago
by
abdullah
🚩 Report: Not working
1
#1 opened 4 months ago
by
mkami2828
[bot] Conversion to Parquet
#1 opened 4 months ago
by
parquet-converter
could not convert string to float: ' Exception
1
#1 opened 5 months ago
by
saddamHGFace
[bot] Conversion to Parquet
#1 opened 4 months ago
by
parquet-converter
Upload Screenshot_2024-09-03-09-22-08-34_1c337646f29875672b5a61192b9010f9.jpg
#2 opened 4 months ago
by
BuradaHanish
Why the images are not added to the dataset ?
#2 opened 5 months ago
by
MohamedRashad
Audio Chat Demo ?
2
#1 opened 5 months ago
by
MohamedRashad