211
GPT-4 is not getting worse
GPT-4 may have declined in some areas, but it hasn’t systematically done so. My tests were shit, and never proved much of anything.
New
The latest blog posts to take off on Hacker News. 789 posts.
212
Optimizing LLMs from a Dataset Perspective
Practical guide to improving LLM finetuning with better instruction datasets, covering data curation, prompt-output pairs, synthetic data, and experiment ideas.
213
Against LLM Maximalism
LLMs are not a direct solution to most of the NLP use-cases companies have been working on. They are extremely useful, but if you want to deliver reliable software you can improve over time, you can't just write a prompt and call it a day. Once you're past prototyping and want to deliver the best system you can…
214
Stable Audio: Fast Timing-Conditioned Latent Audio Diffusion
Stable Audio represents the cutting-edge audio generation research by Stability AI’s generative audio research lab, Harmonai. We continue to improve our model architectures, datasets, and training procedures to improve output quality, controllability, inference speed, and output length.
215
Lessons from YC AI Startups
The evolution of the AI landscape, and YC's Copilot era.
216
Simulating History with ChatGPT
The Case for LLMs as Hallucination Engines
217
A Multi-Level View of LLM Intentionality
Prompted by Keith Frankish's recent streamed discussion of LLM intentionality on YouTube , there's a particular idea I wanted to share which...
218
Nvidia’s AI supremacy is only temporary
Nvidia is an amazing company that has executed a contrarian vision for decades, and has rightly become one of the most valuable corporations on the planet thanks to its central role in the AI revol…
219
Intuitively Understanding Harris Corner Detector
If you ever tried to learn how the Harris corner detection algorithm works, you might have noticed that the process is not intuitive at all. First, you start with an energy function, approximate it using Taylor approximation, get a matrix from that, then find the eigenvalues of that matrix, etc. But when you come to…
220
A LLM+OLAP Solution
The exploration of a LLM+OLAP solution is a bumpy journey, but phew, it now works well for the Tencent case, and they're writing down every lesson learned to share with you.
221
Training and aligning LLMs with RLHF and RLHF alternatives
I frequently reference a process called Reinforcement Learning with Human Feedback (RLHF) when discussing LLMs, whether in the research news or tutorials.
222
We built an AI-powered Magic the Gathering card generator
Cohere's platform was used to create an AI that would generate playable Magic The Gathering cards based only on a given name of each card.
223
Persimmon-8B
We’re open-sourcing Persimmon-8B, the most powerful fully permissively-licensed language model with <10 billion parameters.
224
Mojo is available for local download
Since our launch of the Mojo programming language on May 2nd, more than 120K+ developers have signed up to use the Mojo Playground and 19K+ developers actively discuss Mojo on Discord and GitHub. Today, we’re excited to announce the next big step in Mojo’s evolution: Mojo is now available for local download …
225
Microsoft will assume liability for legal copyright risks of Copilot
With customers ask whether they can use Microsoft’s Copilot services without worrying about copyright claims, we are providing a straightforward answer: yes, you can, and if you are challenged on copyright grounds, we will assume responsibility for the potential legal risks involved.
226
227
Gaussian splatting is pretty cool
SIGGRAPH 2023 just had a paper “ 3D Gaussian Splatting for Real-Time Radiance Field Rendering ” by Kerbl, Kopanas, Leimkühler, Drettakis, and it looks pretty cool! Check out their website, source code repository, data sets and so on (I should note that it is really, really good to see full source and full data sets…
228
Playing with genetic algorithms in Python
Edit: This post made it to the front page of Hacker News! You can find the discussion here.
229
Remembering Doug Lenat and his quest to capture the world with logic
Stephen Wolfram shares memories of Doug Lenat, early AI pioneer and developer of the CYC system. Their discussions around AI & logic and the potential combined uses of their respective discoveries.
230
Can LLMs learn from a single example?
We’ve noticed an unusual training pattern in fine-tuning LLMs. At first we thought it’s a bug, but now we think it shows LLMs can learn effectively from a single example.
231
LLM Python/CLI tool adds support for embeddings
LLM is my Python library and command-line tool for working with language models. I just released LLM 0.9 with a new set of features that extend LLM to provide tools …
232
Refact Code LLM: 1.6B LLM for code that reaches 32% HumanEval
Today we're introducing Refact LLM: 1.6B code model with infill real-time code completion (including fill-in-the-middle(FIM) capability) and chat.
233
234
Doug Lenat has died
AI has lost a giant
235
AI model weight providers should not police uses, no matter how awful they are
The Electronic Frontier Foundation, a digital civil liberties advocacy group, recently published a good article entitled “ISPs Should Not Police Online Speech—No Matter How Awful It Is.” 1 It argues that providers within the internet “stack” should refrain from censoring traffic according to their own values. That…
236
237
Fine-tuning GPT-3.5-turbo for natural language to SQL
Tutorial and comparison with RAG methods
238
Teaching with AI
We’re releasing a guide for teachers using ChatGPT in their classroom—including suggested prompts, an explanation of how ChatGPT works and its limitations, the efficacy of AI detectors, and bias.
239
Long Live the 'GPU Poor' – Open-Source AI Grants
a16z announces its Open Source AI Grant program, which will support a small group of open source developers through grant funding.
240
High-Speed AI Drone Overtakes World-Champion Drone Racers
Remember when IBM’s Deep Blue won against Gary Kasparov at chess in 1996, or Google’s AlphaGo crushed the top champion Lee Sedol at Go, a much more complex game, in 2016? These competitions where machines prevailed over human champions are key milestones in the history of artificial intelligence. Now a group of…