This is the code for the paper Donkii: Can Annotation Error Detection Methods Find Errors in Instruction-Tuning Datasets?
pip install -r requirements
bash run_experiments.sh
This is the code for the paper Donkii: Can Annotation Error Detection Methods Find Errors in Instruction-Tuning Datasets?
pip install -r requirements
bash run_experiments.sh