Irem Ergün

3.7K posts

Irem Ergün banner
Irem Ergün

Irem Ergün

@irombie

•mle @cohere; LLM posttraining •i write code&songs&stories •🇹🇷🇬🇧 tweets •proud Bilkenter •chasing legacy✨ •views are mine, not my employer's. rt≠endorsement

Riverside, CA Katılım Aralık 2020
554 Takip Edilen2.3K Takipçiler
Farving🙆⭐️
Farving🙆⭐️@FarvingCo·
Memory loss doesn’t start in your brain. 2 days ago, Stanford published a study in Nature identifying the exact 3-step pathway: Step 1: Aging shifts your gut bacteria (P. goldsteinii overgrows) Step 2: These bacteria produce fatty acids that trigger IL-1β inflammation in the gut Step 3: Inflammation silences your vagus nerve → hippocampus stops forming memories They transferred old gut bacteria into young mice. The young mice became forgetful. They stimulated the vagus nerve in old mice. Cognitive performance matched young animals. The lead researcher called the gut “a remote control for the brain.” (DOI: 10.1038/s41586-026-10191-6): This flips everything we assumed about aging brains. You’re not losing your edge because of age. You’re losing it because your gut is inflamed and your brain can’t hear its own body anymore. Fix the gut. Get sharp again. Brain fog. Belly fat. Fatigue that sleep doesn’t fix. These aren’t 3 problems. They’re 1 broken pathway. Take my free 2-minute gut assessment quiz to find out where yours is broken. Comment LEAK.
Farving🙆⭐️ tweet media
English
56
294
1.6K
62.6K
Irem Ergün retweetledi
Hugging Models
Hugging Models@HuggingModels·
Meet Turkish-Gemma-9b-T1: a powerful, specialized LLM fine-tuned for Turkish language tasks. It's not just another translation model. It's a full conversational AI that understands, reasons, and generates fluent Turkish text. This is a big deal for Turkish NLP.
Hugging Models tweet media
English
14
82
944
61.4K
Irem Ergün
Irem Ergün@irombie·
I am excited to announce that I will be a mentor for Expedition Tiny Aya! Together, we will come up with and work on an impactful research project, building off of the Tiny Aya models. I literally cannot wait until next week for the kickoff 🤩🤩 If you are interested in multilingual NLP research, go and sign up 🤏🏻🔥🫵
Cohere Labs@Cohere_Labs

We’re officially kicking off Expedition Tiny Aya, a global, mentor-supported open build challenge built on the Tiny Aya model — and we’d love to see you there.

English
2
3
39
3.6K
Irem Ergün retweetledi
Kiran Vodrahalli
Kiran Vodrahalli@kiranvodrahalli·
I'm happy to announce that we are open sourcing the MRCR v2 datasets we report on Gemini releases! github.com/google-deepmin…
Kiran Vodrahalli@kiranvodrahalli

Happy to share Michelangelo (arxiv.org/abs/2409.12640), a long-context reasoning benchmark which measures performance beyond needle tasks up to arbitrary context lengths and remains challenging for frontier models. Stay tuned for more Michelangelo evals to come!

English
2
9
60
24.5K
Irem Ergün retweetledi
Cohere Labs
Cohere Labs@Cohere_Labs·
Tiny Aya shows what smaller models can do. It improves on previous Aya releases and outperforms models at similar size proving that smart multilingual design can rival larger models. This shows that focused multilingual research beats brute-force scaling—achieving more with less.
Cohere Labs tweet media
English
2
9
51
21.6K
Irem Ergün
Irem Ergün@irombie·
I will be speaking at my alma mater next Tuesday (Feb 17). Super stoked, see you there!! ❤️‍🔥 Thanks @bilkentai for organizing!
Irem Ergün tweet media
English
2
2
14
1.1K
Irem Ergün retweetledi
Saurabh Dash
Saurabh Dash@TheyCallMeMr_·
Calling it compute rich/poor and not Jensen’s Inequality is a missed opportunity
English
1
8
38
6.9K
Ben Lang
Ben Lang@benln·
Cafe Cursor is coming up in these cities: • Ankara - 1/31 • Cluj - 1/31 • Medellín - 2/7 • Melbourne - 2/14 • Warsaw - 2/16 • Krakow - 2/18 • Copenhagen - 2/19 • Santiago - 2/21 • Manta - 2/24 • Paris - 2/25 • Milan - 2/26 • Hanoi - 2/28 • Quito - 2/28
Română
163
35
1.2K
395.2K

Keşfet

@FarvingCo @mert_xai @sarahookr @mziizm @bilkentai @elonmusk @BarackObama @taylorswift13