PromptPapers

Why we included this project

PromptPapers is a reading list of roughly sixty-five papers on prompt-based tuning for large language models, maintained by researchers at THUNLP who also built the OpenPrompt toolkit. The papers are grouped into sections that move from broad overviews through pilot work and the basics of template and verbalizer design to analyses, improvements, and specializations, which gives a newcomer a sensible path into the literature. Each entry links to the original paper and, where available, its code, with badges indicating the model family, the prompt-learning technique involved, and the task being studied. For anyone getting oriented in this area, a curated, actively maintained bibliography with pointers to implementations beats assembling your own list from scattered searches. The maintainers accept pull requests, so the collection keeps absorbing new work from the community.