Singh Man Singh
kitsingh.bsky.social
Singh Man Singh
@kitsingh.bsky.social
Don’t let the bastards grind you down: What they don’t want is happy, healthy minorities living their best life— do just that!!
February 25, 2025 at 9:02 PM
Give me good tips on how to speed up the passage of time…need to get to 2029 asap
January 22, 2025 at 12:43 AM
When you are Republican, they let you grab them by the healthcare. #georgiaAccess
January 7, 2025 at 4:07 AM
Georgia Access is Nightmare fuel

I WAS FINE WITH HEALTHCARE MARKETPLACE. THEY JUST BROKE SOMETHING THAT DIDNT NEED FIXING.
January 6, 2025 at 11:25 PM
Dealing with #GeorgiaAccess has been a nightmare.

My provider hasn’t received any updates about my new insurance plan, and now I’m stuck without coverage. Georgia Access shows a false confirmation of enrollment, but my provider confirmed they never received anything.
January 6, 2025 at 11:23 PM
American people are goldfish. Public opinion can be shifted with a few tweets about minorities
January 6, 2025 at 3:17 PM
Reposted by Singh Man Singh
OpenScholar, a retrieval-augmented language model (LM) designed to answer user queries by first searching for relevant papers in the literature and then generating responses grounded in those sources #DL #AI #ML #DeepLearning #ArtificialIntelligence #MachineLearning
GitHub - AkariAsai/OpenScholar: This repository includes the official implementation of OpenScholar: Synthesizing Scientific Literature with Retrieval-augmented LMs.
This repository includes the official implementation of OpenScholar: Synthesizing Scientific Literature with Retrieval-augmented LMs. - GitHub - AkariAsai/OpenScholar: This repository includes the...
buff.ly
November 24, 2024 at 4:31 AM
Bunny butt Friday
November 23, 2024 at 5:33 AM
Reposted by Singh Man Singh
Bluesky is amazing because I can actually follow what the f*ck is going on in the world again on a social media app. Finally.
November 23, 2024 at 4:58 AM
Reposted by Singh Man Singh
How do LLMs learn to reason from data? Are they ~retrieving the answers from parametric knowledge🦜? In our new preprint, we look at the pretraining data and find evidence against this:

Procedural knowledge in pretraining drives LLM reasoning ⚙️🔢

🧵⬇️
November 20, 2024 at 4:35 PM
@anthropic.com Claude shows that Constitution AI was the best approach
November 23, 2024 at 4:25 AM