Ryan Lowe
Ryan Lowe is an AI researcher who co-wrote the 2022 InstructGPT paper, which fine-tunes GPT-3 with human feedback. His early work includes The Ubuntu Dialogue Corpus, a dataset of almost 1 million multi-turn dialogues that he co-wrote in 2015. In 2017 he co-authored Towards an Automatic Turing Test, which won an Outstanding Paper award at ACL. In 2020 he took part in Learning to summarize with human feedback, a NeurIPS paper written with eight others. As of October 2026, he is on the core team of the Meaning Alignment Institute.
- Now at
- Meaning Alignment Institute
- Works
- 31
- Citations
- 7,317
- h-index
- 23
Publication numbers from OpenAlex, as of 15 October 2026.
Timeline
-
2015
The Ubuntu Dialogue Corpus
In 2015 he co-writes The Ubuntu Dialogue Corpus with Nissan Pow, Iulian Serban and Joelle Pineau, a dataset of almost 1 million multi-turn dialogues.
-
2017
Multi-agent actor-critic
He co-authors Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments with Yi Wu, Aviv Tamar, Jean Harb, Pieter Abbeel and Igor Mordatch, a NeurIPS paper from 2017.
-
2017
Towards an Automatic Turing Test
Towards an Automatic Turing Test: Learning to Evaluate Dialogue Responses, which he co-authors, wins an Outstanding Paper award at ACL 2017.
-
2019
Retrospectives workshop
In 2019 he is one of the people listed for Retrospectives: A Venue for Self-Reflection in ML Research, a NeurIPS workshop, with Yoshua Bengio and Joelle Pineau.
-
2020
Learning to summarize with human feedback
In 2020 he writes Learning to summarize with human feedback with eight others, a NeurIPS paper.
-
2022
The InstructGPT paper
In 2022 he co-authors Training language models to follow instructions with human feedback, which fine-tunes GPT-3 with human feedback and calls the results InstructGPT.
-
As of October 2026
Core team at Meaning Alignment
By October 2026 he is on the core team of the Meaning Alignment Institute.
Sources and credits
Photo: We have no published photo of Ryan Lowe yet. Send us a free-licensed photo.
- Ryan Lowe
- [1506.08909] The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems
- Retrospectives: A Venue for Self-Reflection in ML Research
- Learning to summarize with human feedback
- Training language models to follow instructions with human feedback
- [2203.02155] Training language models to follow instructions with human feedback
- Meaning Alignment Institute