Companies specifically focusing on developing LLMs for code generation. There are the obvious ones, but I also know a couple of startups who are laser-focused on this.
@rasbt
-
LLMs struggle with TikZ figures but code generation training improves automation
By
–
Yes, LLM also still struggle with Tikz-generated figures etc. I can imagine though that with all the focus on code-generation training, this will be easier to automate.
-
Apple’s Cloud Infrastructure Choices for AI and Storage Services
By
–
Interesting point. They still use Google Cloud for their iCloud storage service with end-to-end encryption (except UK); so I am surprised they wouldn't consider using similar cloud compute services for their AI workloads until they have their own hardware up and running.
-
Concern about the misuse of the term distillation
By
–
I mean, look what they've done to the term "distillation" in the last few weeks
-
LLMs Will Easily Generate Educational Videos with Code-Based Animations
By
–
Actually, I think this will be the easiest kind of educational video to be generated by LLMs. That's because I believe (pls correct me if I'm wrong) the animations are code generated.
-
HP acquisition impact on Apple stock price analysis
By
–
Don't know, but HP bought them for $116 million, which is a 0.003% of Apple's stock price. So if the announcement about the delay results in more than a 0.01% drop in the stock price, maybe the acquisition to get it out sooner wouldn't have been so bad?
-
Reasoning LLMs: Multiple Implementation Approaches and Research
By
–
Not for now as there are ~100 different ways to implement reasoning. I've written about some of them here: https://
magazine.sebastianraschka.com/p/understandin
g-reasoning-llms
… and plan to follow-up with the latest research in a bit on my blog.
I think Unsloth is a great tool btw! -
LLM Training Easier Than OS and App Integration Challenges
By
–
sure, but the LLM is probably the easier part given that they already have trained a few; and pre-training+post-training usually takes ~3-6 months.
I thought the challenge is more on the integration with the OS and apps side. -
Machine Learning and Statistics: The Inference Terminology Confusion
By
–
I'm used to it by now. Back when I was a statistics professor, I always had to start my lectures by clarifying that "inference" in machine learning has a different meaning than in stats…otherwise chaos ensued.
-
Terminology Update: Test Set and Test Time Rebranding to Evals and Inference
By
–
test set and test time -> evals and inference