121
Duolingo Max, a learning experience powered by GPT-4
Duolingo’s newest subscription, Duolingo Max, offers a powerful AI-backed learning experience.
Most cited
Posts that commenters keep linking to, ranked by how many different people have cited them in Hacker News comments. 155 posts.
122
Prompt engineering vs. blind prompting
"Prompt Engineering" emerged from the growth of language models to describe the process of applying prompting to effectively extract information from language models, typically for use in real-world applications.
123
Let's try to understand AI monosemanticity
Inside every AI is a bigger AI, trying to get out
124
ChatGPT in an iOS Shortcut: Smart HomeKit Voice Assistant
Ever since I tried ChatGPT and GPT-3, everything else feels painfully dumb and useless: Siri, Alexa, Google Home and all other “smart”…
125
Accelerating Generative AI with PyTorch II: GPT, Fast
This post is the second part of a multi-series blog focused on how to accelerate generative AI models with pure, native PyTorch. We are excited to share a breadth of...
126
LLM Powered Autonomous Agents
Building agents with LLM (large language model) as its core controller is a cool concept. Several proof-of-concepts demos, such as AutoGPT, GPT-Engineer and BabyAGI, serve as inspiring examples. The potentiality of LLM extends beyond generating well-written copies, stories, essays and programs; it can be framed as a…
127
OpenAI's plans according to Sam Altman removed at OpenAI's request
Last week I had the privilege to sit down with Sam Altman and 20 other developers to discuss OpenAI’s product plans. Sam was remarkably open. The discussion touched on practical developer issues as well as bigger-picture questions related to OpenAI’s mission and the societal impact of AI. Here are the key takeaways.
128
I’m a doctor: Here’s what I found when I asked ChatGPT to diagnose my patients
ChatGPT recently passed the U.S. Medical Licensing Exam, but using it for a real-world medical diagnosis would quickly turn deadly.
129
Chat Notebooks
Wolfram expands Notebooks, integrating LLM functionality into the new chat cell and Chat Notebook. Stephen Wolfram explains how chat-enabled and chat-driven versions work.
130
What are embeddings?
A deep-dive into machine learning embeddings.
131
The Transformer Family
Many new Transformer architecture improvements have been proposed since my last post on “The Transformer Family” about three years ago. Here I did a big refactoring and enrichment of that 2020 post — restructure the hierarchy of sections and improve many sections with more recent papers. Version 2.0 is a superset of…
132
NotebookLM: An AI Notebook
We’re rolling out NotebookLM, an experimental offering from Google Labs to summarize information, complex ideas and brainstorm new connections.
133
Re-implementing LangChain in 100 lines of code
LangChain has become a tremendously popular toolkit for building a wide range of LLM-powered applications, including chat, Q&A and document search. In this blogpost I re-implement some of the novel LangChain functionality as a learning exercise, looking at the low-level prompts it uses to create these higher level…
134
Gzip beats BERT? Part 2: dataset issues, improved speed, and results
Summary : Update to my last post analyzing the gzip+knn paper. Some of the datasets have heavily contaminated train/test tests. I show some ways to significantly speed-up the gzip method. I present more results of my "fair" reimplementation .
135
136
Will scaling work?
Data bottlenecks, generalization benchmarks, primate evolution, intelligence as compression, world modelers, and other considerations
137
Behind the scenes of Sound ID in Merlin – Identify birds using your phone (2021)
What is Sound ID? Today we announced one of our biggest breakthroughs—Sound ID, a new feature in the Merlin Bird ID app—and a major leap forward in sound identification and machine learning
138
Paving the way to efficient architectures: StripedHyena-7B
StripedHyena-7B is an open-source model built beyond the Transformer architecture, matching Llama-2 and Mistral 7B while running faster on long sequences.
139
Many options for running Mistral models in your terminal using LLM
Mistral AI is the most exciting AI research lab at the moment. They’ve now released two extremely powerful smaller Large Language Models under an Apache 2 license, and have a …
140
Fuyu-8B: A multimodal architecture for AI agents
We’re open-sourcing Fuyu-8B - a small version of the multimodal model that powers our product.
141
AI and Trust
I trusted a lot today. I trusted my phone to wake me on time. I trusted Uber to arrange a taxi for me, and the driver to get me to the airport safely. I trusted thousands of other drivers on the road not to ram my car on the way. At the airport, I trusted ticket agents and maintenance engineers and everyone else who…
142
GPT-4 gets a B on my quantum computing final exam
Update (April 14): On the suggestion of my TA Justin Yirka, I gave GPT-4 problem 2 from the final exam, along with its own answers and my and Justin’s grading comments, and asked it whether i…
143
Emergent abilities of large language models
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.
144
Improving mathematical reasoning with process supervision
We’ve trained a model to achieve a new state-of-the-art in mathematical problem solving by rewarding each correct step of reasoning (“process supervision”) instead of simply rewarding the correct final answer (“outcome supervision”). In addition to boosting performance relative to outcome supervision, process…
145
146
Llama 2 is about as factually accurate as GPT-4 for summaries and is 30X cheaper
Using Anyscale Endpoints, we compared Llama 2 7b, 13b and 70b vs. OpenAI's GPT-3.5-turbo and GPT-4 for accuracy and cost.
147
Programming AIs worry me
For some inane reason, Github classifies me as a "major open source maintainer", which means I get a free copilot subscription.1 I've been using it for a...
148
Diffusion with Offset Noise: Finetuning SD to generate very dark or light images
Historical research record. Fine-tuning against a modified noise, enables Stable Diffusion to generate very dark or light images easily.
149
DuckDuckGo Launches DuckAssist
A new feature that generates natural language answers to search queries using Wikipedia.
150
Leaked Google document: “We Have No Moat, And Neither Does OpenAI”
SemiAnalysis published something of a bombshell leaked document this morning: Google “We Have No Moat, And Neither Does OpenAI”. The source of the document is vague: The text below is …