multilingual-alignment

Tag

Cards List
#multilingual-alignment

Aligned in Form, Not in Meaning: The Comprehension - Containment Decoupling of LLM Safety in Low-Resource Bangla Derogatory Speech

arXiv cs.CL · 6d ago Cached

This arXiv paper audits five frontier LLMs on native Bangla derogatory speech, finding that safety alignment fails to generalize to low-resource languages — models comprehend and generate unsafe content at high rates despite high-resource alignment. The authors propose a 'comprehension–containment decoupling' and show that reasoning and persona framing further break down safety filters.

0 favorites 0 likes
← Back to home

Submit Feedback