Technology 2 min read

New MIT Automated System can Update Wikipedia Articles

Tero Vesalainen /

Tero Vesalainen /

Scientists at MIT have created an automated system that can rewrite outdated Wikipedia articles — a task that human editors currently perform.

Millions of articles on Wikipedia require constant editing to reflect new information. Whether its article expansion, major rewrites, or routine modification, the manual process can be time-consuming.

Luckily, contributors across the globe volunteer their time to make these edits. But they may not need to do that in the future.

In a paper at the AAAI Conference, MIT researchers described a new automated text-generating system. The AI system identifies and rewrites outdated sentences in Wikipedia articles in a human-like fashion.

One of the lead author and a Ph.D. student in the Computer Science and Artificial Intelligence Laboratory (CSAIL), Darsh Shah, said:

“Instead of hundreds of people working on modifying each Wikipedia article, then you’ll only need a few, because the model is helping or doing it automatically. That offers dramatic improvements in efficiency.”

Here’s how it works.

Using an Automated System to Rewrite Outdated Sentences in Wikipedia Articles

Human editors can type an unstructured sentence with updated information into an interface.

In turn, the system would search the online encyclopedia to identify the appropriate page and outdated sentence. Then, it would replace the sentence while keeping the language similar to how humans write and edit.

Tons of other bots can make automatic Wikipedia edits. For example, some systems can drop narrowly defined information into a pre-defined template.

So, why is MIT’s new automated system a big deal?

According to Shah, the new system addresses a more challenging AI problem. Unlike the other bots’ rule-based tasks, the MIT model’s job involves considering the contradictory parts in two sentences and generating a coherent piece of text.

Aside from rewriting Wikipedia articles, the automated system is also useful for other text-generating applications.

For example, the model automatically synthesized sentences in a popular fact-checking dataset that helped reduce bias. What’s more, the researchers didn’t have to collect additional data manually.

This way, the performance improves for automatic fact-verification models that train on the dataset for, say, fake news detection,” says co-lead author, Tal Schuster.

The researchers point out that we could get a fully automated system in the future. Such a model would search the web for the latest information and rewrite corresponding Wikipedia articles.

Read More: New Deep Learning Tool can Find Bugs in Software Updates

First AI Web Content Optimization Platform Just for Writers

Found this article interesting?

Let Sumbo Bello know how much you appreciate this article by clicking the heart icon and by sharing this article on social media.

Profile Image

Sumbo Bello

Sumbo Bello is a creative writer who enjoys creating data-driven content for news sites. In his spare time, he plays basketball and listens to Coldplay.

Comments (0)
Most Recent most recent
share Scroll to top

Link Copied Successfully

Sign in

Sign in to access your personalized homepage, follow authors and topics you love, and clap for stories that matter to you.

Sign in with Google Sign in with Facebook

By using our site you agree to our privacy policy.