This repository curates academic papers and algorithms for offline reinforcement learning, organized into sections covering surveys, theoretical methods, benchmarks, applications, and off-policy evaluation techniques. Maintained by researchers at Cornell University and Hanjuku-kaso Co., Ltd.