Every blog post that took off on Hacker News, ranked by the votes of its best submission. 3,092 posts.
When I first started writing software, developers had to implement most of the algorithms they used; yes, hardware vendors provided libraries, but the culture was one of self-reliance (except for maths functions, which were technical and complicated).
Judea Pearl and Dana Mackenzie sent me a copy of their new book, “The book of why: The new science of cause and effect.”
When code execution really is all you need. Exploring what happens when an AI agent can only write and run code.
A professor at Stanford once said, If you really want to impress your friends and confound your enemies, you can invoke tensor products… People run in terror from the $ \otimes$ symbol. He was explaining some aspects of multidimensional Fourier transforms, but this comment is only half in jest; people get confused by…
RAG poisoning attack walkthrough: an attacker injects fabricated documents into a ChromaDB knowledge base and the LLM reports false financials as fact. Fully reproducible, 100% local, no GPU required.
Lossless compression of a target model's KV cache by up to 4×, using a cheaper predictor model to drive an arithmetic coder.
Large language models are valuable research assistants, especially when they refuse to follow instructions.
Debugging with intelligence via probabilistic inference Xu et al., ICSE’18 Xu et al. have built a automated debugger that can take a single failing test execution, and with minimal interactio…
Design matters more than ever in consumer AI, as long as you have money to play.
Isn't there somebody you forgot to ask?
Or, how well can GPT-4 simulate an acid trip in 1963?
How can we use familiar design patterns to constrain the unwieldy state space of large language models?
Common misconceptions about the complexity in robotics vs AI
AI agents are good at doing one thing at a time. Give Claude a focused task and it performs. But real work isn’t one task. It’s a tree of tasks with dependencies, parallelism, and context that needs to flow between them.
Experiments in cognitive architecture.
We present ObamaNet , the first architecture that generates both audio and synchronized photo-realistic lip-sync videos from any new text. Contrary to other published lip-sync approaches, ours is only composed of fully trainable neural modules and does not rely on any traditional computer graphics methods. More…
A series on automatic differentiation in Julia. Part 1 provides an overview and defines explicit chain rules.
The uncanny resemblance between double descent and a 50 year old theory from psychology
A journey into making Stack Overflow job search results relevant to you
Today, tech folk are scrambling to change their workflows to meet newly inflated 5X productivity quotas, while getting pummeled under the co...
If your initial reaction to reading that and seeing that is some variation of "ughhh" or rolling your eyes or "fuck this guy" congrats. You are normal.
Dell sent me two of their GB10 mini workstations to test: In this blog post, I'll cover the base system, just one of the two nodes. Cluster testing is ongoing, and I'll cover things like AI model training and networking more in depth next year, likely with comparisons to the Framework Desktop cluster and Mac Studio…
“It’s not that hard to build a fully functioning, code-editing agent.” Thorsten Ball
Machine-learning generated content has major implications on the Metaverse, because it brings the marginal cost of production to zero.
AIs limited to pure computation (Tool AIs) supporting humans, will be less intelligent, efficient, and economically valuable than more autonomous reinforcement-learning AIs (Agent AIs) who act on their own and meta-learn, because all problems are reinforcement-learning problems.
Neural Radiance Field training can be accelerated through the use of grid-based representations in NeRF's learned mapping from spatial coordinates to colors and volumetric density. However, these grid-based approaches lack an explicit understanding of scale and therefore often introduce aliasing, usually in the form…
