1 paper · 1 filter
Vaishak Narayanan, Prabin Raj KP, Saifudheen Nouphal
Current evaluation benchmarks for question answering (QA) in Indic languages often rely on machine translation of existing English datasets. This approach suffers from bias and ina…