Official code for ACL 2023 paper: Multilingual LLMs are Better Cross-lingual In-context Learners with Alignment
To run the corresponding experiments run following:-
Please follow the instructions in dataset_processing.ipynb for preprocessing the data
run dataset_creation.py to generate setups
run MARC, CLS and Hateval usin MARC_run.py, cs_run.py and hateval_run.py