AI safety researcher

Leo Gao

Google Scholar dblp

Risky behavior 1Defenses 1Evaluation and benchmarks 1

Latest AI safety papers

On this site: 4 papers

March 2025

Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
Mar 14Risky behaviorarXiv ↗

December 2023

Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
Dec 14DefensesarXiv ↗

March 2023

GPT-4 Technical Report
Mar 15Evaluation and benchmarksarXiv ↗

October 2022

Scaling Laws for Reward Model Overoptimization
Oct 19arXiv ↗

Authorship follows dblp and the author's own Google Scholar profile; only AI safety papers on this site are listed.