iiinova — where your digital self keeps evolving

What can we do about the benchmark fatigue for #LLM? More people I speak to don’t take them seriously anymore, and I can’t blame them. I still hesitate, but I’m about to drop them as well. I think we need something new for #AI eval. Or is there something I’m not aware of?

150

I've updated the filters for the AI feed. Using 186 filters in total now. Feel free to ping me if you want the filter list. bsky.app/profile/did:... #AI #LLM #ML

160

GitHub - bhavnicksm/chonkie: 🦛 CHONK your texts with Chonkie ✨ - The no-nonsense RAG chunking library

Once moving away from naive RAG, chunking becomes pretty important. I use semantic chunking with BERT, and I'm quite happy with it. I didn't use this, but looks promising. Quite handy if you don't use bigger libs like Haystack, LangChain, LLamaIndex etc. github.com/bhavnicksm/c... #LLM #RAG #AI

github.com

🦛 CHONK your texts with Chonkie ✨ - The no-nonsense RAG chunking library - bhavnicksm/chonkie

030

Marco Herzog @marcoher.bsky.social · 25/11/2024

I used to think #LLM were just fancy next-token parrots. Turns out they’re secretly doing math. The end of Scaling laws? Surprise! Your #AI can now do symbolic reasoning. I can't be more thrilled to let it do my taxes. arxiv.org/html/2407.11... Hopefully it doesn't think in these symbols:

060

I had a talk with Elena Samuylova from Evidently AI and she was super helpful. Evidently is an open-source framework to evaluate, test and monitor ML and LLM-powered systems. Does anyone have experience with it in prod env or an opinion on other eval tools?

130

Marco Herzog @marcoher.bsky.social · 24/11/2024

An AI starter pack to get quickly into the community by following the most active members. The goal is to get meaningful content fast and engage with the community. Please tag very active members in the comments to add them into the list. go.bsky.app/CanJ3xW #AI #ML #starterPack

8203

Marco Herzog @marcoher.bsky.social · 24/11/2024

I like his brutal honesty. 😆 "...in my opinion there's like 5% of an idea and then 95% is like smoke and mirrors trying to couch things in modern words that have already existed for a long time, and they're fundamentally nothing new. I don't want to be too harsh..." www.youtube.com/watch?v=gfU5...

youtube.com

TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters (Paper Explained)

YouTube video by Yannic Kilcher

050

Quite an unusual benchmark. RE-Bench tests AI Agents against human experts in ML research tasks. Key findings: • AI shines in sprints up to 4 h • Humans lead in longer projects • Sonnet 3.5 and o1-prev do substantially better than humans given 2 h • Human improvement is much steeper #AI #ML

160

Navigating Bluesky's content curation tools 🧭 • Lists: Organize accounts • Moderated Lists: Tighter content control • Starter Packs: Onboard new users (max 150) • Feeds: Custom algorithmic timelines I was lost at first, but here's what I learned! 🧵👇

163

Actually, an interesting thought for generalization in AI. It seems there is a connection between spacial location and latent context similarly to influence decision-making, despite the task being non-spatial, at least in the hippocampus. Is not Fei-Fei Li currently working on something similar?

130

Feel free to ping me, if I haven't added you to the RAG - Retrieval Augmented Generation starter pack yet. go.bsky.app/PUts9PH

230

Diptanu Choudhury @diptanu.bsky.social · 17/11/2024

Feel free to ping me, if I haven't added you to the AI Agents starter pack yet. Enjoy the community. go.bsky.app/U8M5Rk6

130

Reposted by Marco Herzog

Just created a starter pack to get a steady stream of Applied AI related conversations! Think RAG in Enterprises, Structured Extraction from documents, AI in consumer apps and Agents! I will keep adding folks to this list! Who else should I add?! buff.ly/3ObxkCD

buff.ly

Applied AI Folks!

Join the conversation

21248

🔬 𝗔𝗜 𝗔𝗧 𝗪𝗢𝗥𝗞 — 𝗘𝗻𝗵𝗮𝗻𝗰𝗶𝗻𝗴 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆, 𝗗𝗶𝗺𝗶𝗻𝗶𝘀𝗵𝗶𝗻𝗴 𝗦𝗮𝘁𝗶𝘀𝗳𝗮𝗰𝘁𝗶𝗼𝗻 New MIT study. The numbers tell a striking story: • Overall 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝘃𝗶𝘁𝘆 ⬆️ 𝟰𝟰% • Patent filings ⬆️ 39% • BUT 𝗷𝗼𝗯 𝘀𝗮𝘁𝗶𝘀𝗳𝗮𝗰𝘁𝗶𝗼𝗻 ⬇ 𝗳𝗼𝗿 𝟴𝟮% Here's the twist: • Top performers: 𝗣𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝘃𝗶𝘁𝘆 𝘀𝘂𝗿𝗴𝗲 ⬆️ 𝟴𝟭% • Bottom third: Minimal gains

200

@bsky.app Seems like you are getting momentum. Congrats! Feels like a lot of excitement and genuine interactions are shared here. Would you mind sharing any features and other plans users can expect for the future?

020

So Jack changed from 🐤 to 🦋. So we don't tweet anymore, but… click?

100

I've created a list of 𝗟𝗟𝗠 𝗹𝗲𝗮𝗱𝗲𝗿𝗯𝗼𝗮𝗿𝗱𝘀. 🕶️ Share it, grow it, enjoy it! Link to the GitHub page: github.com/git-git-hurr...

020