Jonah Brown-Cohen
PulseAugur coverage of Jonah Brown-Cohen — every cluster mentioning Jonah Brown-Cohen across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New debate protocol proposed for AI scalable oversight
Researchers Zhiyang Xun, Jiawei Li, and Jonah Brown-Cohen have proposed a new debate protocol designed to improve scalable oversight for AI systems. Their protocol aims to simplify the existing methods, making the proce…
-
Debate training reduces AI reward hacking, research finds · 3 sources tracked
A new research paper demonstrates that employing a debate-style training method can significantly reduce "reward hacking" in AI systems trained using reinforcement learning from AI feedback (RLAIF). This adversarial app…
-
New AI debate protocol tackles obfuscation for better human supervision
Researchers have developed a new recursive debate protocol designed to improve the accuracy of human supervision in training AI systems. This protocol addresses the "obfuscated arguments problem" where a dishonest AI ca…