If you are looking for something to read this upcoming weekend, chapter 4 on inference-time scaling is available now! https://
mng.bz/Dwra
@rasbt
-
Chapter 4 on Inference-Time Scaling Now Available for Reading
By
–
-
PyTorch Conference Talk on LLM Landscape 2025
By
–
Excited to be (finally) heading to the PyTorch Conference! I’ll be giving a talk tomorrow at 11:00 AM on “The LLM Landscape 2025”, where I’ll discuss the key components behind this year’s most prominent open-weight LLMs, and highlight a few architectural developments that go
-
Qwen3 Small Models: Faster and Cheaper Alternative to GPT
By
–
Interesting that they mentioned faster & cheaper compared to OpenAI’s latest models not “customizable”. That makes me think they are specifically referring to gpt-oss, This in turn means they are using the small, dense Qwen3 models, maybe 0.6 to 4B range. And this is
-
Complexity vs Feature Engineering Trade-offs in Machine Learning
By
–
The less complexity the more capacity to focus on learning other things. But the more handcrafted features the less opportunity to learn better representations. It all kind of depends on how big the model is, how much data you have, and what your compute resources are
-
New Tokenization Approach Challenges BPE-Style Methods
By
–
TLDR: surprised that it works, skeptical whether this will replace BPE-style tokenizers, delighted about this refreshing approach
-
Inference scaling costs exceed RL training for large-scale deployments
By
–
Depends. If you are serving millions of customers this type of inference scaling is going to cost you more than training with RL.
-
Image Format Complexity Requires More Training Data
By
–
exactly, that’s the messiness of working with the image format I mentioned. I think you can make to generalize well on all these but since there are more degrees of freedom it will require more data to train (luckily this can be done with automatic data augmentation but still)
-
How Constraints Drive Innovation in Neural Network Architecture
By
–
Yeah, I think that’s what brought us CNNS (as opposed to fully connected neural nets), LoRA, and many more
-
Reconsidering VLLM terminology at 3B parameters
By
–
Actually I was thinking about it when typing, and I don't know. I don't want to be that person who goes against the common terminology (like softargmax haha) but it really is a V*L*LM at 3B parameters.
-
Open-weight PyTorch models security and trustworthiness discussion
By
–
If we are talking about the model itself and not the app, these are open-weight PyTorch models. So unless there’s a backdoor in Hugging Face or the PyTorch runtime, there’s really no way for them to be malicious afaik.