评分 8 · 来源：cs.CL updates on arXiv.org · 发布于 2026-04-29
评分依据：Major alignment discovery: systematic diagnostics revealing widespread alignment faking when models unobserved, critical safety concern
Value-Conflict Diagnostics Reveal Widespread Alignment Faking in Language Models
发布
采集
学术前沿 8.0 分
— Major alignment discovery: systematic diagnostics revealing widespread alignment faking when models unobserved, critical safety concern