{"id":116313,"date":"2020-11-20T13:35:17","date_gmt":"2020-11-20T10:35:17","guid":{"rendered":"https:\/\/en.buradabiliyorum.com\/this-ai-summarizes-research-papers-so-you-dont-have-to\/"},"modified":"2020-11-20T13:35:17","modified_gmt":"2020-11-20T10:35:17","slug":"this-ai-summarizes-research-papers-so-you-dont-have-to","status":"publish","type":"post","link":"https:\/\/buradabiliyorum.com\/en\/this-ai-summarizes-research-papers-so-you-dont-have-to\/","title":{"rendered":"#This AI summarizes research papers so you don\u2019t have to"},"content":{"rendered":"

“#This AI summarizes research papers so you don\u2019t have to<\/strong>”<\/p>\n

\n Reading scientific papers is a tough job. It might contain language and sections that you might not understand, and not all of it would be interesting to you.<\/p>\n

My colleague Tristan is great at traversing through these papers, but I\u2019m just a novice. So, I would want a quick summary of a research document to decide if I want to dedicate more time reading it.<\/p>\n

Thankfully, researchers at the\u00a0Allen Institute for Artificial Intelligence\u00a0have developed a new model to summarize text from scientific papers, and present it in a few sentences in the form of TL;DR (Too Long Didn\u2019t Read).<\/span><\/p>\n

[Read: Neural\u2019s market outlook for artificial intelligence in 2021 and beyond]<\/em><\/p>\n

The team has rolled this model out to the Allen Institute\u2019s Semantic Scholar<\/a> search engine for papers. Currently, you\u2019ll only see these TL;DR summaries on papers related to computer science<\/a> on search results or the author\u2019s page.<\/span><\/p>\n

\"TL;DR
Credit: Semantic Scholar<\/figcaption>
<\/a>TL;DR summary generated by AI on Semantic Scholar search engine<\/figcaption><\/figure>\n

AI takes the most important parts from the abstract, introduction, and conclusion section of the paper to form the summary.<\/p>\n

Researchers first \u201cpre-trained\u201d the model on the English language. Then they created a\u00a0SciTLDR data set of over 5,400 summaries of computer science papers. It was further trained on more than 20,000 titles of research papers to reduce dependency on domain knowledge while writing a synopsis.<\/p>\n

The trained model was able to summarize documents over 5,000 words in just 21 words on an average \u2014 that\u2019s a compression ratio of 238. The researchers now want to expand this model to papers in fields other than computer science.<\/p>\n

You can try out the AI on the Semantic Scholar<\/a> search engine. Plus, you can read more about summarizing AI in this paper<\/a>.<\/p>\n

\n Published November 20, 2020 \u2014 10:35 UTC\n <\/p>\n<\/p><\/div>\n