4711
Step 5 Preview: Advancing the Pareto Frontier
阶跃星辰于2023年4月成立，以“智能阶跃，十倍每个人的可能”为使命。阶跃星辰坚定自研超级模型，积极布局算力、数据等关键资源，发挥算法和人才优势，已完成 Step-1 千亿参数语言大模型和 Step-1V 千亿多模态大模型的研发，在图像理解、多轮指令跟随、数学能力、逻辑推理、文本创作等方面性能达到业界领先水平。
Top posts
Every blog post that took off on Hacker News, ranked by the votes of its best submission. 6,334 posts.
4712
The Math Behind GANs (2020)
Generative Adversarial Networks refer to a family of generative models that seek to discover the underlying distribution behind a certain data generating process. This distribution is discovered through an adversarial competition between a generator and a discriminator. As we saw in an earlier introductory post on…
4713
Current architectural best practices for LLM applications
In this guide, I will share the standard architecture for data-informed language model applications and explain forthcoming improvements in knowledge retrieval.
4714
4715
SVD Image Compression
Singular Value Decomposition (SVD) is a fundamental concept in linear algebra, and it is particularly important in the field of machine learning for tasks such as dimensionality reduction, data compression, and noise reduction.
4716
Demystifying Text Data with the Unstructured Python Library
In the world of data, textual data stands out as being particularly complex. It doesn’t fall into neat rows and columns like numerical data does. As a side project, I’m in the process of developing my own personal AI assistant. The objective is to use the data within my notes and documents to answer my questions. The…
4717
RETRO is fast
When I first read Google’s RETRO paper, I was skeptical. Sure, RETRO models are 25x smaller than the competition, supposedly leading to HUGE savings in training and inference costs. But what about the new trillion token “retrieval database” they added to the architcture? Surely that must add back some computational…
4718
Experimenting with LLMs to Research, Reflect, and Plan
Also, shortcomings in document retrieval and how to overcome them with search & recsys techniques.
4719
Recurrent Neural Network Tutorial for Artists (2017)
Recurrent Neural Network Tutorial for Artists
4720
Interpreting neurons in an LSTM network
A few months ago, we showed how effectively an LSTM network can perform text transliteration .
4721
ChatGPT Saved My Life (no, seriously, I'm writing this from the ER)
How using AI as a bridge when doctors aren't available can improve patient-to-doctor communications in real time emergencies
4722
4723
LLMs tell bad jokes because they avoid surprises
LLMs suck at comedy, art, journalism, research, and science for the same fundamental reason
4724
Grok 4.1
Grok 4.1 is now available to all users on grok.com, 𝕏, and the iOS and Android apps. It is rolling out immediately in Auto mode and can be selected explicitly as “Grok 4.1” in the model picker.
4725
What's working for YC companies since the AI boom
The goal of this analysis was to understand what’s working for YC companies by looking at which ones have raised Series A rounds.
4726
The AI Productivity Gap
Why the productivity gains from AI are still small.
4727
Marc Andreessen on Investing and Tech
Transcript of interview on how tech changes the world
4728
Run Qwen3.8 27B locally: real numbers from my Mac Studio
Qwen3.8 27B benchmarked on a Mac Studio M3 Ultra: Ollama and llama.cpp numbers, the 1-bit quant tested, RAM math, and the hardware that actually fits.
4729
Waymo and Hyundai enter multi-year, strategic partnership
Today, Waymo and Hyundai Motor Company announced they have entered into a multi-year, strategic partnership.
4730
AMÁLIA and the future of European Portuguese LLMs
Thoughts on the new technical report from AMÁLIA: The Open Source LLM for European Portuguese
4731
Notes on gotchas while migrating 35kb preprompts from Opus to self-hosted Ollama
Maybe you’re a Claude code/codex user diligently avoiding uploading personal data to LLM providers. Is it possible that the most valuable information isn’t your data- but the metadata about your sessions? The intuitions you apply in coming up with ways to coax the ai into solving problems might actually be special…
4732
Is Lovable getting monetization wrong?
The current (and future) monetization strategies of "Shopify for SaaS"
4733
Duet AI for Google Workspace Now Available
We’re making Duet AI for Google Workspace available to organizations around the world and announcing new AI capabilities, including automatic summaries and action items in Meet.
4734
Something weird is happening with LLMs and Chess
are they good or bad?
4735
Getting Started with Mistral-7b-Instruct-v0.1
Fast and Portable Llama2 Inference on the Heterogeneous Edge.
4736
FreeWilly 1 and 2, two new open-access LLMs
Stability AI and its CarperAI lab are proud to announce FreeWilly1 and its successor FreeWilly2, two powerful new, open access, Large Language Models (LLMs). Both models demonstrate exceptional reasoning ability across varied benchmarks.
4737
Bypassing Gemma and Qwen safety with raw strings
How a Single Function Call Gates Safety Alignment in Gemma, Qwen, and Other Open-Source LLMs
4738
In the land of LLMs, can we do better mock data generation?
"Lorem ipsum is latin, slightly jumbled, the remnants of a passage from Cicero's _de Finibus_ 1.10.32, which begins 'Neque porro quisquam est qui dolorem ipsum quia dolor sit amet, consectetur, adipisci velit...' [There is no one who loves pain itself, who seeks after it and wants to have it, simply because it is…
4739
Using Deep Learning to Predict the Olfactory Properties of Molecules
Posted by Alexander B Wiltschko, Senior Research Scientist, Google Research Smell is a sense shared by an incredible range of living organisms, a...
4740
NanoGPT Speedrun Frontier
Compare 153 autonomous runs from 18 frontier models on the nanoGPT optimizer speedrun.