LLM Digest
Subscribe

Story

arxiv_llm_reliability · Nov 18, 2026 · paper

Source brief

Can Weight-Space Programming Alter Learned Ethical Priorities?

arxiv.orgNov 18, 2026
original source linked

In brief

We investigate whether a weight edit can alter how a model trades off honoring a commitment against benefiting more people. In a concept-supervised text model, we alter fitted choice weights so that the benefit required to override a formal pledge is larger than for an ordinary p

Continues in

Weight-Space Programming — open the evidence trace →

Continue reading

Read the original at arxiv.org →Open in live feedRead that day’s brief

Earlier in this thread 4 items