4501
A Short Chronology of Deep Learning for Tabular Data
Occasionally, I share research papers proposing new deep learning approaches for tabular data on social media, which is typically an excellent discussion…
Top posts
Every blog post that took off on Hacker News, ranked by the votes of its best submission. 6,334 posts.
4502
Flare: Clojure Dynamic Neural Net Library
I wrote Flare, a Dynamic Neural Net library in Clojure.
4503
FlashAttention: Fast Transformer training with long sequences
Transformers have grown deeper and wider, but training them on long sequences remains difficult. The attention layer at their heart is the compute and memory bottleneck: doubling the sequence length would quadruple the runtime and memory requirements.
4504
Introduction to Bert
Unless you have been out of touch with the Deep Learning world, chances are that you have heard about BERT — it has been the talk of the town for the last one year. At the end of 2018 researchers …
4505
Computational predictions of protein structures associated with COVID-19
The scientific community has galvanised in response to the recent COVID-19 outbreak, building on decades of basic research characterising this virus family. Labs at the forefront of the outbreak response shared genomes of the virus in open access databases, which enabled researchers to rapidly develop tests for this…
4506
"AI, no ads please": 4 words to wipe out $1T
Introducing a series on AI vs. ads
4507
4508
Providing ChatGPT to the U.S. federal workforce
Today, OpenAI for Government is announcing a new partnership with the U.S. General Services Administration (GSA) to launch a transformative initiative. For the next year, ChatGPT Enterprise will be available to the entire federal executive branch workforce at essentially no cost.
4509
Rise of generative AI will be comparable to the rise of CGI in the early 90s
My chat with Cristóbal Valenzuela, Cofounder & CEO of Runway .
4510
A Model of a Mind
This article explains a simple model of how minds might work. I’m motivated by the success of AI-based language models to look at the future of digital minds. I’ll present a conceptual data-flow architecture that can account for several key features of minds: the ability to initiate actions (agency), learning…
4511
I turned my unused Kindle into a photo frame with GPT-6 Astra
I turned my unused Kindle into a photo frame with an AI agent.
4512
How AI text watermarking works
A gentle visual guide to how a statistical mark hides inside generated text, and what erases it.
4513
Converting Codebases with LLMs
Leverage LLMs with context like code structure, libraries, and screenshots to streamline code conversion tasks and accelerate development.
4514
4515
Cerebras Inference now 3x faster: Llama3.1-70B breaks 2,100 tokens/s
Cerebras powers the world's fastest AI inference on the biggest wafer chip. Cerebras CS-4 delivers up to 30x faster inference than GPUs.
4516
What my dad taught me about AI coding in the 90s
It was a summer holiday, sometime in the late 1990s. I was a young kid, around 8 or so. My dad had taught me the rules of chess before and we played from time to time. But this time was different: As we were about to start he stood up and walked away from the board, sitting down in a chair across the room and facing…
4517
Researchers run high-performing LLM on the energy needed to power a lightbulb
UC Santa Cruz researchers show that it is possible to eliminate the most computationally expensive element of running large language models, called matrix multiplication, while maintaining performance.
4518
Gradients Are the New Intervals
At the New England Symposium on Graphics , James Tompkin compared graphics researchers to magpies: they're easily distracted by shiny objects and pretty renderings.
4519
The Google Assistant is going global
Over the past year and a half, the Google Assistant has grown from being available on just one device in one language to across many types of devices, including speakers…
4520
4521
Unifying the CUDA Python Ecosystem
For years, a Python developer who needed a GPU had two realistic choices: Learn NVIDIA CUDA C++ well enough to write an extension, set up a build toolchain, and maintain bindings back to Python…
4522
Midjourney is alemwjsl
aadil pickle's internet home
4523
vLLM large scale serving: DeepSeek 2.2k tok/s/h200 with wide-ep
How vLLM reaches 2.2k tokens per second per H200 for DeepSeek-style MoE serving with Wide-EP, async scheduling, dual-batch overlap, disaggregated serving, CUDA
4524
Doing a Machine Learning PhD While Working in Japan
An American MEXT scholar shares the pros and cons of doing a PhD in Japan, from funding and part-time work to university life and career options.
4525
Einsum Is All You Need – Einstein Summation in Deep Learning (2018)
Tim Rocktäschel — Co-Founder and Chief Scientist of Recursive; Professor of AI at UCL. Artificial General Intelligence, Open-Endedness, and Recursive Self-Improvement.
4526
Everything I've learned so far about running local LLMs
Over the past month I’ve been exploring the rapidly evolving world of Large Language Models (LLM). It’s now accessible enough to run a LLM on a Raspberry Pi smarter than the original ChatGPT (November 2022). A modest desktop or laptop supports even smarter AI. It’s also private, offline, unlimited, and…
4527
Predict the future with Machine Learning
Learn Machine Learning (ML) and be ahead of the pack in the Software industry. Machine Learning is a sub-field of Artificial Intelligence (A.I), which is heavily used in modern software systems. Machine Learning algorithms can improve software (a robot) and it's ability to solve problems through gaining experience and…
4528
A Eureka machine that thinks like nature and explores what AI cannot
IISc is the premier institute for advanced scientific and technological research and education in India.
4529
Making a Neural Synthesizer Instrument
In a previous post, we described the details of NSynth (Neural Audio Synthesis), a new approach to audio synthesis using neural networks. We hinted at furthe...
4530
LLM Neuroanatomy II: Modern LLM Hacking and Hints of a Universal Language?
In Part 1, I described how duplicating a block of seven middle layers in Qwen2-72B — no weight changes, no training — produced the #1 model on the HuggingFace Open LLM Leaderboard. The method, which I called RYS (Repeat Your Self), was discovered using nothing but hard math probes and EQ-Bench on a pair of RTX 4090s.