close
close

Association-anemone

Bite-sized brilliance in every update

Duke professors launch Inquisite, an AI personal assistant that promises to make research easier
asane

Duke professors launch Inquisite, an AI personal assistant that promises to make research easier

Two Pratt School of Engineering professors have launched a personal AI research engine whose machine learning models are being adapted for use in academia.

the program, Closedwas developed primarily by Jon Reifschneider, Executive Director of the AI ​​for Product Innovation Master’s Program, with assistance from Pramod Singh, Executive Director-in-Residence in the Engineering and Professional Graduate Programs. It was officially released to the public on October 17.

Inquisite works similarly to other language models and search engines such as ChatGPT or Google, but specializes in conducting research on complex topics. Its main attraction is the promise to provide users with answers generated from trusted sources.

Unlike other models, Inquisite provides citations for all the sources it uses in its answers, as well as relevance and reliability rankings for each source.

“If you go to Google and type in your question, you’re going to get a bunch of stuff, and half of it is going to be people’s blogs about what they think … or maybe misinformation or maybe conflicting information and you’re not sure which one to rely on,” he said. Reifschneider said. “Inquisite does this for you and prioritizes using trusted sources.”

According to Reifschneider, Inquisite identifies credible sources by assigning trust scores to academic websites. These scores depend on factors such as the number of citations and whether the source has been published.

Reifenschneider and Singh started the project to help researchers find important information in their fields and optimize their time.

“(A researcher’s) results would be as good as the first part, which is the search part,” Singh said. “Your search must be good and bring up relevant information.”

After talking with potential users, the professors realized that Inquisite’s improved search return feature could be used by a wider audience than just academics.

“For the vast majority of people, they either don’t have the time to read (research papers) or they don’t have the ability, because they’re not a trained expert in the field.” Reifenschneider said. “What we’re ultimately trying to do is make this kind of reliable information easily accessible.”

To launch and market their product, members of Office of Translations and Marketing provided Reifenschneider and Singh with support to create the technology.

After the initial launch, Inquisite receives feedback from users and uses it to grow. So far, such input has produced a feature that allows users to track the questions they ask the model.

Singh and Reifenschneider are already planning the tool to access and judge non-academic sources and incorporate more factors into the trust scores to increase their reliability.

“We are also now working on integrating other types of signals, such as item pitch,” Reifenschneider said. “Is it written in a factual, objective manner, or is it written in a sensational manner? Does the source itself cite other sources as references in it, or is it just free text that someone sat down and wrote based on their own opinions?

Ultimately, the founders hope to use Inquisite to make information more accessible.

“We believe this can help democratize and revolutionize research,” Singh said.


Winston Qian
| Health/Science Editor

Winston Qian is a Pratt student and health/science editor for the news department.