OpenAI's ChatGPT has released a new update to give users more control over how it responds. A.I. companies have agreed to voluntary safeguards to manage the risks associated with their technology. "Secrets of RLHF in Large Language Models Part I: PPO" introduces a new approach to reinforcement learning with human feedback, which is important for large language models. Finally, "SPRINT: A Unified Toolkit for Evaluating and Demystifying Zero-shot Neural Sparse Retrieval" introduces a new paradigm in retrieval called neural sparse retrieval, and a toolkit called SPRINT to evaluate and compare different models.
Contact:Â Â sergi@earkind.com
Timestamps:
00:34 Introduction
01:48Â Custom instructions for ChatGPT
03:20Â Pressured by Biden, A.I. Companies Agree to Guardrails on New Tools
05:17Â llama2.c Repository by Andrej Karpathy
06:31 Fake sponsor
08:27Â Secrets of RLHF in Large Language Models Part I: PPO
10:19Â Provably Faster Gradient Descent via Long Steps
11:42Â SPRINT: A Unified Toolkit for Evaluating and Demystifying Zero-shot Neural Sparse Retrieval
13:49 Outro
