iiinova — where your digital self keeps evolving

Tal Schuster

@talschuster.bsky.social

133 followers 102 following 4 posts

Research Scientist at Google DeepMind working on Gemini and Adaptive Compute LLMs

Posts Replies Media

Reposted by Tal Schuster

Jeff Dean @jeffdean.bsky.social · 25/03/2025

🥁Introducing Gemini 2.5, our most intelligent model with impressive capabilities in advanced reasoning and coding. Now integrating thinking capabilities, 2.5 Pro Experimental is our most performant Gemini model yet. It’s #1 on the LM Arena leaderboard. 🥇

3421866

Reposted by Tal Schuster

jacobaustin123.bsky.social @jacobaustin123.bsky.social · 04/02/2025

Making LLMs run efficiently can feel scary, but scaling isn’t magic, it’s math! We wanted to demystify the “systems view” of LLMs and wrote a little textbook called “How To Scale Your Model” which we’re releasing today. 1/n

39429

Reposted by Tal Schuster

Simone Scardapane @sscardapane.bsky.social · 18/12/2024

*Relaxed Recursive Transformers* by @talschuster.bsky.social et al. Converts pre-trained transformers to a more efficient version by turning blocks of layers into a single layer which is iterated. Lots of interesting tricks! arxiv.org/abs/2410.20672

152

Tal Schuster @talschuster.bsky.social · 10/12/2024

Will be at NeurIPS. Reach out if you're interested in discussing adaptive compute in LLMs or other topics

000

Tal Schuster @talschuster.bsky.social · 06/12/2024

New Gemini model grabs first place in all domains. Happy one year anniversary Gemini team!

021

Reposted by Tal Schuster

Simon Willison @simonwillison.net · 06/12/2024

Google just released a new Gemini model - gemini-exp-1206 I upgraded my llm-gemini plugin to support it and then got the best result yet for my "Generate an SVG of a pelican riding a bicycle" benchmark simonwillison.net/2024/Dec/6/g...

Blue sky, green grass, bicycle looks good, bird riding it is almost recognizable as a pelican

714210

Tal Schuster @talschuster.bsky.social · 29/11/2024

Dear algorithm, I would like to view: 70% new ML and LLM research and cool results. 10% funny videos with cute animals. 5% sports (but no spoilers if I'm planning to watch the reply later). 5% travel and life hacks. 5% general tech. 5% random. Regards,

182