ICML 2024 DL4C workshop focusing on human-centered coding agents features research from Stanford, Meta, and Anthropic

Sessions cover vulnerability detection and scaling RL using off-policy prefixes