Skip to content
Thursday, 19 November 2026
Tenesys AI News
Subscribe

ETH Zurich

Latest news, models and how they changed.

Latest news

Research· Important· 🧪 Worth Testing

LLM Behavior Control via Input Optimization

Researchers introduced ActGCG, a method to replicate the effects of activation steering—typically requiring internal model access—by optimizing input suffixes. The technique successfully reproduces specific behaviors like language steering through external text inputs alone.

arXiv cs.CL