Ethan Perez
Leads the alignment team at Anthropic; co-author of the RAG, red-teaming, and Constitutional Classifiers work
Ethan Perez leads the alignment team at Anthropic. Perez's site says Perez led the team that developed Constitutional Classifiers, safeguards trained on synthetic data generated from a natural-language constitution to defend against universal jailbreaks. Perez is first author of "Red Teaming Language Models with Language Models", which uses one language model to generate test cases that find harmful behavior in another, and a co-author of the retrieval-augmented generation (RAG) paper.1234
Work
Constitutional Classifiers (external site: arxiv.org)Published research
Last-listed author of the 2025 Anthropic paper introducing Constitutional Classifiers; Perez's site says Perez led the team that developed them.21
Red Teaming Language Models with Language Models (external site: arxiv.org)Published research
First author of the 2022 paper, which automatically generates test cases with one language model to find cases where a target model behaves harmfully, instead of relying only on hand-written tests.31
Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (external site: arxiv.org)Published research
Co-author of the 2020 paper that combines a pre-trained language model with a retrievable non-parametric memory for knowledge-intensive tasks.41
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.