1 paper
Jophin John, Michael Hoffmann, Piroska Lendvai +1
Post-training quantisation reduces the memory cost of large language models, but it remains unclear whether compression preserves language skills of regional varieties as reliably…