Alaa El-Nouby
@alaa_nouby
Followers
741
Following
915
Media
17
Statuses
218
Research Scientist at @Apple. Previous: @Meta (FAIR), @Inria, @MSFTResearch, @VectorInst and @UofG
Paris, France
Joined August 2023
๐๐ผ๐ฒ๐ ๐ฎ๐๐๐ผ๐ฟ๐ฒ๐ด๐ฟ๐ฒ๐๐๐ถ๐๐ฒ ๐ฝ๐ฟ๐ฒ-๐๐ฟ๐ฎ๐ถ๐ป๐ถ๐ป๐ด ๐๐ผ๐ฟ๐ธ ๐ณ๐ผ๐ฟ ๐๐ถ๐๐ถ๐ผ๐ป? ๐ค.Delighted to share AIMv2, a family of strong, scalable, and open vision encoders that excel at multimodal understanding, recognition, and grounding. (๐งต)
4
27
153
RT @EmmanuelMacron: Consistent with its historic commitment to a just and lasting peace in the Middle East, I have decided that France willโฆ.
0
20K
0
RT @tokenbender: we missed a banger paper in the grok4/k2 drop noise guys. these guys .> look for optimal ways to select data mixes to geโฆ.
0
76
0
RT @AggieInCA: If you are at attending ICML today, consider checking out Samaraโs poster on the role of sparsity in MoEs at 11 AM PDT. Postโฆ.
0
2
0
Deciding which data mixture to use has always been such a crucial part for nailing a good pre-training recipe. Check out this paper, led by @PierreAblin , @MustafaShukor1 and the team at Apple MLR, providing a principled way for selecting optimal data mixture weights!.
We propose new scaling laws that predict the optimal data mixture, for pretraining LLMs, native multimodal models and large vision encoders !. Only running small-scale experiments is needed, and we can then extrapolate to large-scale ones. These laws allow 1/n ๐งต
0
4
58
RT @MustafaShukor1: We propose new scaling laws that predict the optimal data mixture, for pretraining LLMs, native multimodal models and lโฆ.
0
48
0
RT @CMHungSteven: @CVPR is around the corner!!.Join us at the Workshop on T4V at #CVPR2025 with a great speaker lineup (@MikeShou1, @jw2yanโฆ.
0
19
0
RT @MustafaShukor1: The Worldwide @LeRobotHF hackathon is in 2 weeks, and we have been cooking something for youโฆ .Introducing SmolVLA, aโฆ.
0
82
0
RT @gm8xx8: SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics. PAPER: .
0
22
0
RT @j_foerst: Hello World: My team at FAIR / @metaai (AI Research Agent) is looking to hire contractors across software engineering and ML.โฆ.
docs.google.com
We are looking for contractors. If you have a track record of ML-Ops and / or SWE excellence and are looking to work with us on a contracting basis, please fill in below.
0
23
0
RT @demishassabis: Me and the Egyptian King ๐ best player in the world - 47 G/As, totally unreal season. Let me know if you ever fancy a gaโฆ.
0
155
0
๐คฏ.
0
0
5
RT @paulg: I don't have to tell you what happened to these three boys. You already know. How awful is that?
0
9K
0
RT @danbusbridge: Iโve been curious about how early vs late-fusion multimodal approaches compare in controlled conditions. Great to see thiโฆ.
0
8
0
RT @alaa_nouby: We have been thinking a lot about how to train truly native multimodal models:. (1) what arch to use (early-fusion, late-fuโฆ.
0
27
0
RT @AkshatS07: Excited to see further studies into early fusion vs late fusion models, in particular a great analysis into multimodal MoEโsโฆ.
0
8
0
RT @anaralabs: Apple just broke the scaling laws for image models. Imagine creating Ghibli art, but 10x faster.
0
56
0