Research

AI Neural Networks Research

Formalizing Explanations of Neural Network Behaviors

This video is from Simons Institute. Existing research on mechanistic interpretability usually tries to develop an informal human understanding of “how a model works,” making it hard to evaluate research results and raising concerns about scalability. Meanwhile formal proofs of model properties seem far out of reach both in theory and practice. In this talk […]

Read More
AI Large Language Models Research

‘Show Your Working’: ChatGPT Performance Doubled w/ Process Rewards (+Synthetic Data Event Horizon)

This video is from AI Explained is a few weeks old, but it’s amazing look at the state of the art LLMs. At least what state of the art was last month. I have not only read the Let’s Verify step by Step paper released less than 24 hours ago, I have combed the release […]

Read More
AI Research

DeepMind’s New AI: History In The Making!

Two Minute Papers examines the paper “Faster sorting algorithms discovered using deep reinforcement learning.”

Read More
Large Language Models Research

Memory in LLM Applications

This video is from Weights & Biases. This talk by Harrison Chase of Langchain will focus on how “memory” in the context of LLM applications. Memory is most often discussed in the context of chatbots so it will start with an overview of conversational memory. This can vary from simple (buffer of recent messages) to […]

Read More
AI Large Language Models Microsoft Research

Orca: The Model Few Saw Coming

The first model set to be opensourced that actually comes close to ChatGPT, and is just 13B (that’s small enough for a laptop). The 51 page report from Microsoft was just released. By imitating the logic and explanations of GPT 4 (and using GPT 3.5 as an assistant), as well as by training on diverse […]

Read More
AI Hardware Research

NVIDIA Is Simulating 100,000 Hair Strands!

Two Minute Papers highlights the paper “Interactive Hair Simulation on the GPU using ADMM.”

Read More
AI Large Language Models Research

Unlock 🚀 900% Logic & Reasoning Improvement with New Prompt (GPT-4)

In this video, Matthew Berman reviews a recently published paper from Google Deepmind and Princeton that describes a new prompting technique to achieve far greater accuracy on logic and reasoning problems with ChatGPT (GPT-4). I’ll also show you how to install and run code implementing this technique. Links: Paper – https://arxiv.org/abs/2305.10601 GitHub – https://github.com/kyegomez/tree-of-thoughts

Read More
AI Research

Can Artificial intelligence Decode Brain Activity into Dialogue?

Scientists at UT Austin conducted a study where they created a 3D view of a person’s mind and used artificial intelligence to decode brain activity into dialogue.

Read More