Data sources: paper metadata, TLDR summaries and citation counts from Semantic Scholar; arXiv metadata from arXiv (thank you to arXiv for use of its open access interoperability); further metadata from Crossref and the LessWrong / Alignment Forum API; Hacker News comments and stories via the HN Search API by Algolia, each linked back to Hacker News. Lab posts are listed by title and the site's own short description, linking to the original on anthropic.com, openai.com and deepseek.com.
Papers, posts and comments belong to their authors and publishers. This site only links to them and hosts no PDFs. Themes are assigned automatically and may be wrong. Not affiliated with or endorsed by Semantic Scholar, arXiv, Crossref, Hacker News, Y Combinator, Anthropic, OpenAI or DeepSeek.
Want your comment, username or paper removed, or spotted an error? Get in touch.