Question: Are there any search engines that use machine learning to build answers from a massive library of academic literature?

Epsilon is an AI-powered search engine geared specifically for scientific discovery. It's got features like Investigate that offers answers with inline citations and Synthesize that generates longer summaries based on vetted papers. Epsilon uses GPT-4 to summarize text and has been adopted by more than 30,000 researchers around the world.

Another good option is Lumina, which uses more than 15 models to construct answers from a corpus of more than 1 million journal papers. It's designed to offer fast and authoritative research results, so it's good for researchers who have to digest a lot of academic literature as quickly as possible.

Elicit is another tool geared for researchers, with features like paper search, data extraction and theme detection. It can even automate tasks like systematic reviews and keep up with the latest research in areas like biomedicine and machine learning.

Last, Semantic Scholar is a free AI-powered research service that lets scholars search for and read relevant scientific papers from a database of more than 219 million papers. It's got powerful search, brief summaries and tools to cite papers and manage your research, all without any subscription costs.