conda create -n feedbackmt python=3.10.0
conda activate feedbackmt
cd src/LMFlow
pip3 install -e .
cd ../../
pip3 install -r requirements.txt
conda install mpi4py
Download from Google drive.
unzip data.zip
See training_scripts
.
python3 src/inference_sft.py \
--model-name-or-path <model path> \
--inst-file data/instruct_follow.txt \
--lang-pair en-zh \
--input-file <input file> \
--output-file <output file> \
--search-algorithm beam \
--batch 2 \
--seed 0 \
--model-type s2s \ # --model-type s2s for NLLB; --model-type causal for LLAMA-2
--beam 4
src/LMFlow/src/lmflow/pipeline/raft_aligner.py # RAFT/RAFT+ for LLAMA2
src/LMFlow/src/lmflow/pipeline/raft_aligner_t2t.py # RAFT/RAFT+ for NLLB
src/LMFlow/src/lmflow/pipeline/mrt_aligner_t2t.py # MRT/MRT+ for NLLB
@inproceedings{he2024improving,
title = "Improving Machine Translation with Human Feedback: An Exploration of Quality Estimation as a Reward Model",
author={Zhiwei He, Xing Wang, Wenxiang Jiao, Zhuosheng Zhang, Rui Wang, Shuming Shi, Zhaopeng Tu},
booktitle = "Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
year = "2024",
url = "https://aclanthology.org/2024.naacl-long.451",
}
- OptimalScale/LMFlow: The RAFT implementation is based on
LMFlow
. - wxjiao/ParroT: Training and inference scripts are based on
ParroT