Aaditya Ura's Avatar

Aaditya Ura

@aadityaura.bsky.social

ML Researcher | LLMs โšก Healthcare | Developer of OpenBioLLM, Open Medical-LLM Leaderboard, MedMCQA, Med-Halt | I like high-quality datasets & eval ๐Ÿง 

126 Followers  |  23 Following  |  5 Posts  |  Joined: 23.11.2024  |  1.7584

Latest posts by aadityaura.bsky.social on Bluesky

Post image

Last Week in Medical AI: Top Research Papers/Models ๐Ÿ”ฅ
๐Ÿ… (December 15 - December 21, 2024)

Full thread in detail: x.com/OpenlifesciA...
Youtube Link: www.youtube.com/watch?v=SbFp...
Spotify: t.co/QPmdrXuWP9

22.12.2024 01:43 โ€” ๐Ÿ‘ 3    ๐Ÿ” 1    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Post image

How can pre-trained foundation models transform healthcare?

@GoogleAI introduces Health AI Developer Foundations (HAI-DEF), a suite of domain-specific tools and models for diverse medical modalities!

Hereโ€™s what you need to know: ๐Ÿ‘‡๐Ÿงต 1/5

28.11.2024 10:58 โ€” ๐Ÿ‘ 8    ๐Ÿ” 1    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Added :)

27.11.2024 09:40 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

This is a great list! Would love to be added. I work at the intersection of AI x healthcare and developer of the OpenBioLLM model :)

27.11.2024 09:39 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Sure :)

27.11.2024 09:20 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

@pminervini.bsky.social Here we go :)

27.11.2024 09:13 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Preview
Medical AI Join the conversation

For my fellow researchers in AI, Medical, and Healthcare domain:, Here is the Medical AI Startup pack if you are new there
go.bsky.app/PddA2uy

27.11.2024 09:13 โ€” ๐Ÿ‘ 15    ๐Ÿ” 4    ๐Ÿ’ฌ 4    ๐Ÿ“Œ 0
Post image

It's Sunday morning so taking a minute for a nerdy thread (on math, tokenizers and LLMs) of the work of our intern Garreth

By adding a few lines of code to the base Llama 3 tokenizer, he got a free boost in arithmetic performance ๐Ÿ˜ฎ

[thread]

24.11.2024 11:05 โ€” ๐Ÿ‘ 272    ๐Ÿ” 34    ๐Ÿ’ฌ 5    ๐Ÿ“Œ 5
Post image Post image Post image Post image

All the things you need to know to pretrain an LLM at home*!

Gave a workshop at Uni Bern: starts with scaling laws and goes to web scale data processing and finishes training with 4D parallelism and ZeRO.

*assuming your home includes an H100 cluster

19.11.2024 20:35 โ€” ๐Ÿ‘ 77    ๐Ÿ” 9    ๐Ÿ’ฌ 5    ๐Ÿ“Œ 0

@aadityaura is following 14 prominent accounts