A live, daily-updating semantic map of AI research. Papers from arXiv's cs.AI category are fetched, embedded, grouped by an AI taxonomist, augmented with additional sources, and rendered as an interactive 2D scatter plot, making it easy to navigate the research landscape, spot emerging themes, and find related work.
I used to view arxiv's cs.ai papers every day, wading through page upon page of listings. I had no idea which papers were promising or more legit. I created the AI Research Atlas to help me and others stay up to date on the latest in AI research. In addition to the arxiv feed, the site integrates TLDRs from Semantic Scholar, uses Haiku for taxonomy, SPECTER for scientific paper embeddings and grouping, and gathers both paper citation data plus authors' academic ratings (h_indices). With all this info and filtering criteria for it, you can zero in better on the research papers that matter. The Atlas displays papers in both a scatterplot and a grid.
WhereWeMet.Org