Artificial Intelligence can do TL;DR Better than Humans

How will open-source software continue to transform our world? How will AI-human hybrid teams continue to boost productivity and allow GDP and other indicators of quality of life to thrive?

What if AI can help us be smarter, and what does that look like in 2021? Researchers at the Allen Institute for Artificial Intelligence have 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’t Read).

AI can now sum up a research paper in a single sentence and the source code is being made available. Think about it, search engine tools will also evolve. A search engine’s tool for summarizing studies promises easier skim-reading. In a world of headlines, algorithms will make us skim read better too.

The creators of a scientific search engine have unveiled software that automatically generates one-sentence summaries of research papers, which they say could help scientists to skim-read papers faster.

• The free tool, which creates what the team calls TLDRs (the common Internet acronym for ‘Too long, didn’t read’), was activated this week for search results at Semantic Scholar, a search engine created by the non-profit Allen Institute for Artificial Intelligence (AI2) in Seattle, Washington.
• AI takes the most important parts from the abstract, introduction, and conclusion section of the paper to form the summary.
• Preliminary testing suggests that the tool helps readers to sort through search results faster than viewing titles and abstracts, especially on mobile phones, he says.

This is a feel good story of the power of open-source software and AI development. The researchers have made their code freely available, along with a working demo website where anyone can try the tool.

Dan Weld, who manages the Semantic Scholar group at AI2 spoke to Nature about it recently. So how did they do it?

Weld was inspired to create the TLDR software in part by the snappy sentences his colleagues share on Twitter to flag up articles. Like other language-generation software, the tool uses deep neural networks trained on vast amounts of text. The team included tens of thousands of research papers matched to their titles, so that the network could learn to generate concise sentences. 

• The trained model was able to summarize documents over 5,000 words in just 21 words on an average — that’s a compression ratio of 238.
• The team has gathered training examples to improve the software’s performance in 16 other fields, with biomedicine likely to come first.
• TLDR can generate a sentence from a paper’s abstract, introduction and conclusion. Its summaries tend to be built from key phrases in the article’s text, so are aimed squarely at experts who already understand a paper’s jargon. But Weld says the team is working on generating summaries for non-expert audiences.

The NLP revolution is changing how humans experience data thanks to AI. You can try out the AI on the Semantic Scholar search engine. Plus, you can read more about summarizing AI in this paper. Congrats to the researchers at the Allen Institute for Artificial Intelligence (AI2). You can check out the GitHub here.

Skim reading, deep fakes, at this point AI could practically create a TikTok video and you'd think it was real. Meanwhile armies around the world are attempting to create synthetic telepathy, while Apple is making us more cybernetic. While we socially distance during a pandemic, the internet is likely also changing faster than normal.

Liked this article? Follow me and subscribe to the Last Futurist.