dair-ai/ai-papers-of-the-week

This repository is a weekly curated list of notable AI research papers, maintained by the DAIR.AI organization. Each week the team reviews papers published in the field of machine learning and artificial intelligence, then highlights the ones they consider most significant. The lists are organized by year and week, going back several years, so the repository functions as a running archive of what was considered worth reading at each point in time. The content is entirely editorial, meaning there is no code, tool, or application here. It is a collection of links and short descriptions pointing to research papers. If you follow AI research and want a human-filtered shortlist rather than sorting through everything published each week, this is what the repository provides. DAIR.AI also runs a newsletter called NLP News on Substack where the same weekly selections are delivered by email. The repository exists alongside that newsletter as a browsable GitHub-hosted version of the same content. The topics covered span machine learning broadly, with a focus on deep learning, natural language processing, and the kind of research that tends to get attention in the AI community week to week. The repository does not explain or summarize the papers itself, it links to them and names them, so readers still need to go read the papers directly. If you are looking for a way to track what is happening in AI research without building your own reading list from scratch, this repository offers a maintained starting point. The archive covers 2023 onward based on the structure visible in the README.