TabuLM is the first language model pre-trained on Kinyarwanda tabular data

We present TabuLM, the first language model pre-trained on Kinyarwanda tabular data.

http://arxiv.org/abs/2608.26923v1