AI Exchange
timezone
+00:00 GMT
SIGN IN
  • Home
  • Events
  • Content
  • People
  • Messages
  • Channels
  • Help
Sign In
Sign in or Join the community to continue

OpenAI’s InstructGPT: Aligning Language Models With Human Intent

Posted Sep 09
# NLP
# LLM
# GPT
# AI
# OpenAI
Share
SPEAKERS
Long Ouyang
Long Ouyang
Long Ouyang
Research Scientist @ OpenAI

Long Ouyang is a research scientist at OpenAI, where he works on human-in-the-loop machine learning. He has helped build variants of GPT, such as InstructGPT and WebGPT. Previously, he obtained his PhD in cognitive psychology from Stanford.

He's the lead author of, Training language models to follow instructions with human feedback: https://arxiv.org/abs/2203.02155

+ Read More

Long Ouyang is a research scientist at OpenAI, where he works on human-in-the-loop machine learning. He has helped build variants of GPT, such as InstructGPT and WebGPT. Previously, he obtained his PhD in cognitive psychology from Stanford.

He's the lead author of, Training language models to follow instructions with human feedback: https://arxiv.org/abs/2203.02155

+ Read More
Aerin Kim
Aerin Kim
Aerin Kim
Engineering Manager @ Scale AI

Aerin Kim is the Engineering Manager at Scale AI, turning raw data into high-quality training data using machine learning. She is fascinated by the science aspect of training data, the one and only input of deep learning that determines the performance of the model. Prior to Scale, Aerin was a Senior Research Software Engineer at Microsoft where she worked on question answering, semantic parsing and training data generation for AI applications. Before joining Microsoft, Aerin received her Masters degree in Operations Research from the Fu Foundation School of Engineering and Applied Science at Columbia University.

+ Read More

Aerin Kim is the Engineering Manager at Scale AI, turning raw data into high-quality training data using machine learning. She is fascinated by the science aspect of training data, the one and only input of deep learning that determines the performance of the model. Prior to Scale, Aerin was a Senior Research Software Engineer at Microsoft where she worked on question answering, semantic parsing and training data generation for AI applications. Before joining Microsoft, Aerin received her Masters degree in Operations Research from the Fu Foundation School of Engineering and Applied Science at Columbia University.

+ Read More
SUMMARY

Making language models bigger does not inherently make them better at following a user's intent. For example, large language models can generate outputs that are untruthful, toxic, or simply not helpful to the user. In other words, these models are not aligned with their users. In this presentation, Long Ouyang, Research Scientist at OpenAI, and Aerin Kim, Engineering Manager at Scale AI will explore an avenue for aligning language models with user intent on a wide range of tasks by fine-tuning with human feedback.

Join Long, OpenAI Research Scientist, and Aerin, Scale AI Engineering Manager, for a technical presentation followed by a discussion around Long's work with InstructGPT, and audience Q&A.

+ Read More

Watch More

55:22
Posted Oct 06 | Views 13.5K
# TransformX 2021
# Fireside Chat
44:51
Posted Oct 06 | Views 31.2K
# TransformX 2021
# Keynote
See more