This is code for paper "Large-Scale Online Learning for Generative List Recommendation in E-commerce: An Environment Policy Optimization Approach"(SIGIR 2026).
git clone https://github.com/Divpeter/EPO.git
cd EPO
make init
For facilitate the training of the generator, we provide a version of the checkpoints of EPO_evaluator that have been pretrained. Please ensure you have Git LFS installed and fetch the EPO_evaluator model files after cloning the repository.
git lfs pull
Then you need to decompress it.
tar -xzvf ./model/save_model_ad/10/*.tar.gz -C ./model/save_model_ad/10/
python run_reranker.py
Model parameters can be set by using a config file, and specify its file path at --setting_path, e.g., python run_ranker.py --setting_path config. The config files for the different models can be found in example/config. Moreover, model parameters can also be directly set from the command line.
python eval_last_model.py --reload_path='path/to/model/you/trained'