Originally published in together.ai, Sept 11, 2023. Large Language Models (LLMs) have changed the world. However, generating text with them can be slow and expensive. While methods like speculative decoding have been proposed to accelerate the generation speed, their intricate nature has left many in the open-source community hesitant to embrace them. That’s why we’re
Originally published in Pete Warden’s Blog, Nov 11, 2022. I like to think (it has to be!) of a cybernetic ecology where we are free of our labors and joined back to nature, returned to our mammal...
Originally published in Fast.AI, Sept 4, 2023. 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....
Originally published in The Verge, Sept 7, 2023. Microsoft is telling customers it will assume legal responsibility if they get sued for copyright infringement while using the company’s AI Copilot services. In a blog post about an...
Originally posted in The Verge, Aug 24, 2023. Meta said Code Llama will make it easier to finish code. Meta has released a tool called Code Llama, built on top of its Llama 2 large language model, to generate...
Originally published in Meta, Aug 2, 2023. Imagine a professional musician being able to explore new compositions without having to play a single note on an instrument. Or a small business owner adding a soundtrack to their...
Originally published in Amazon Science, Aug 3, 2023. Assessing the absolute utility of query results, rather than just their relative utility, improves learning-to-rank models. When a customer clicks on an item in a list of product-search results,...
Originally published in IEEE Spectrum, July 30, 2023. Five conclusions from an automation expert fresh off a stint with the U.S. highway safety agency. In 2016, just weeks before the Autopilot in his Tesla drove Joshua Brown to...
Originally published in Ars Technica, Aug 1, 2023. Amid competition from TikTok, Meta looks to the next frontier of user engagement. Meta is reportedly developing a range of AI-powered chatbots with different personalities, a move aimed at...
Originally published in Decrypt, July 19, 2023. UC Berkeley researchers found that ChatGPT has not improved over time, and in fact, may have gotten worse. ChatGPT exploded onto the scene late last year, dazzling people with its...
