Valdini Douglace Lemofouet
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
Aligning Large Language Models for Safety in African Languages: A Case Study of Hausa
According to the abstract, Lemofouet, Uzor, Anyanwu and co-authors build HausaSafeLM, a Hausa safety preference dataset of 10,000 prompts, and tune Qwen2.5-7B-Instruct with SFT and DPO. They report the untuned model looks safe by refusal rate only because it generates Hausa poorly.