Story
arxiv_cs_cl · Aug 11, 2026 · paper
Source brief
Actions Speak Louder than Words: Measuring Cross-Lingual Policy Retention in Tool-Using Agents
arxiv.orgAug 11, 2026
original source linked
In brief
When a tool-using agent is given the same task in a different language, does it still take the same steps? Multilingual evaluation rarely asks: it compares final answers and discards the actions. Yet those actions are...
Feed lens
agentevaluation