New paper: value training in LLMs causes unintended side effects on other values

IAugenstein · x · 2026-11-18

Researchers from the University of Copenhagen and University of Cambridge (Siddhesh Pawar, Chen Cecilia Liu, Anna Korhonen, Isabelle Augenstein) released the paper "Unintended Side Effects of Value Training in LLMs".

It adds a measurement lens for "train one value, others drift too" in alignment research.

Original post →

More from Research

Research channel →