Lifeboat Foundation

8 years of cost reduction in 5 weeks: how Stanford’s Alpaca model changes everything, including the economics of OpenAI and GPT 4. The breakthrough, using self-instruct, has big implications for Apple’s secret large language model, Baidu’s ErnieBot, Amazon’s attempts and even governmental efforts, like the newly announced BritGPT.

I will go through how Stanford put the model together, why it costs so little, and demonstrate in action versus Chatgpt and GPT 4. And what are the implications of short-circuiting human annotation like this? With analysis of a tweet by Eliezer Yudkowsky, I delve into the workings of the model and the questions it rises.

Continue reading “The Model That Changes Everything: Alpaca Breakthrough (ft. Apple’s LLM, BritGPT, Ernie and AlexaTM)” »

To effectively tackle everyday tasks, robots should be able to detect the properties and characteristics of objects in their surroundings, so that they can grasp and manipulate them accordingly. Humans naturally achieve this using their sense of touch and roboticists have thus been trying to provide robots with similar tactile sensing capabilities.

A team of researchers at the University of Hong Kong recently developed a new soft tactile sensor that could allow robots to detect different properties of objects that they are grasping. This sensor, presented in a paper pre-published on arXiv, is made up of two layers of weaved optical fibers and a self-calibration algorithm.

“Although there exist many soft and conformable tactile sensors on robotic applications able to decouple the normal force and shear forces, the impact of the size of object in contact on the force calibration model has been commonly ignored,” Wentao Chen, Youcan Yan, and their colleagues wrote in their paper.

A new paper published in the Journal of Medical Internet Research describes how generative models such as DALL-E 2, a novel deep learning model for text-to-image generation, could represent a promising future tool for image generation, augmentation, and manipulation in health care. Do generative models have sufficient medical domain knowledge to provide accurate and useful results? Dr. Lisa C Adams and colleagues explore this topic in their latest viewpoint titled “What Does DALL-E 2 Know About Radiology?”

First introduced by OpenAI in April 2022, DALL-E 2 is an artificial intelligence (AI) tool that has gained popularity for generating novel photorealistic images or artwork based on textual input. DALL-E 2’s generative capabilities are powerful, as it has been trained on billions of existing text-image pairs off the internet.

To understand whether these capabilities can be transferred to the medical domain to create or augment data, researchers from Germany and the United States examined DALL-E 2’s radiological knowledge in creating and manipulating X-ray, computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound images.

With inflation sparking an increase in the cost of repairs, labor and claims, fees for insurance are similarly spiking across the board. Car insurance premiums rose 13.7% nationally over the past year, according to a study from Bankrate.com. Home insurance, meanwhile, climbed 12.1% year-on-year, Policygenius found.

But Jonathan Matus argues that it doesn’t have to be that way. He’s the founder of Fairmatic, a company that’s applying AI to — at least according to him — reduce risk in the car insurance industry.

Matus previously founded Zendrive, a platform that provides insights to enterprises for car insurance underwriting and claims as well as roadside assistance. While Zendrive is focused on insurance for individuals and families, Fairmatic has a more commercial bent — a customer base made up primarily of businesses.

HUMANS in the next 100 years could be part-machine, part-flesh creatures with brain chips and bionic limbs and organs in a vision of “cyborgs” once described by Elon Musk.

Men and women born around 2100 could live in a world very different to ours as humans may be totally connected to the internet and meshed together with artificial intelligence.

Mobile phones would no longer be needed — as everything you now do with your smartphone will now be done with a chip in your brain.

Summary: Text-to-image generation deep learning models like OpenAI’s DALL-E 2 can be a promising new tool for image augmentation, generation, and manipulation in a healthcare setting.

Source: JMIR Publications

A new paper published in the Journal of Medical Internet Research describes how generative models such as DALL-E 2, a novel deep learning model for text-to-image generation, could represent a promising future tool for image generation, augmentation, and manipulation in health care.

The NLP community has recently discovered that pretrained language models may accomplish various real-world activities with the help of minor adjustments or direct assistance. Additionally, performance usually becomes better as the size grows. Modern language models often include hundreds of billions of parameters, continuing this trend. Several research groups published pretrained LLMs with more than 100B parameters. The BigScience project most recently made BLOOM available, a 176 billion parameter model that supports 46 natural and 13 computer languages. The public availability of 100B+ parameter models makes them more accessible, yet due to memory and computational expenses, most academics and practitioners still find it challenging to use them. For inference, OPT-175B and BLOOM-176B require more than 350GB of accelerator RAM and even more for finetuning.

As a result, running these LLMs typically requires several powerful GPUs or multi-node clusters. These two alternatives are relatively inexpensive, restricting the potential study topics and language model applications. By “offloading” model parameters to slower but more affordable memory and executing them on the accelerator layer by layer, several recent efforts seek to democratize LLMs. By loading parameters from RAM just in time for each forward pass, this technique enables executing LLMs with a single low-end accelerator. Although offloading has high latency, it can process several tokens in parallel. For instance, they are producing one token with BLOOM-176B requires at least 5.5 seconds for the fastest RAM offloading system and 22 seconds for the quickest SSD offloading arrangement.

Additionally, many machines lack sufficient RAM to unload 175B parameters. LLMs may be made more widely available through public inference APIs, where one party hosts the model and allows others to query it online. This is a fairly user-friendly choice because the API owner handles most of the engineering effort. However, APIs are frequently too rigid to be used in research since they cannot alter the model’s control structure or have access to its internal states. Additionally, the cost of some research initiatives may be exorbitant, given the current API price. In this study, they investigate a different approach motivated by widespread crowdsourcing training of neural networks from scratch.

Brilliant jerks who nobody actually likes working with will have a tougher time after AI automates much of their work. The future belongs to those with people skills.

Harry’s Review of The Mountain In The Sea by Ray Nayler

Ray Nayler’s recent contemplative sci-fi thriller The Mountain In The Sea offers a first contact story which holds a mirror to our notions of intelligence and responsibility.

Following ecologist Dr Ha Nguyen, The Mountain In The Sea centres upon the recent discovery of an octopus species in the remote Vietnamese islands of the Con Dao archipelago. Questions abound: just how intelligent are these octopuses? Corporations and activists alike become interested. DIANIMA-a vast organisation interesting in automation and artificial intelligence hire Dr Nguyen to determine how valuable the octopus species are to their research. Accompanying her for this job is DIANIMA’s own previous attempt at creating a being that can pass the Turing test, a silicon lifeform known as Evrim.