This repository maintains a curated list of research papers on Language-Model-as-a-Service (LMaaS), a paradigm where users access large pre-trained LLMs through inference APIs without access to model weights. The collection covers methods for adapting black-box LLMs to downstream tasks, including text prompt design, in-context learning, black-box optimization, feature-based learning, and data generation techniques. It is maintained by NLP researchers to facilitate research in efficiently using commercial and closed-source LLMs.