This paper illustrates the architecture and training of Unipa-GPT, a chatbot relying on a Large Language Model, developed for assisting students in choosing a bachelor/master degree course at the University of Palermo. Unipa-GPT relies on gpt-3.5-turbo, it was presented in the context of the European Researchers' Night (SHARPER night). In our experiments we adopted both the Retrieval Augmented Generation (RAG) approach and fine-tuning to develop the system. The whole architecture of Unipa-GPT is presented, both the RAG and the fine-tuned systems are compared, and a brief discussion on their performance is reported. Further comparison with other Large Language Models and the experimental results during the SHARPER night are illustrated.

Voxel51, a startup developing a platform to analyze unstructured data, such as images and videos, has raised $12.5 million in a Series A round led by Drive Capital, with participation from Top Harvest, Shasta Ventures, eLab Ventures and ID Ventures. Founder and CEO Jason Corso tells TechCrunch that the new capital will be put toward further developing the company's platform and doubling the size of Voxel51's team from 13 to 26 employees by year-end. Corso says he, alongside machine learning PhD Brian Moore, created Voxel51 to harness the growing flood of unstructured data in AI and machine learning. A professor at the University of Michigan, Corso says he saw a "critical need" for better software infrastructure to support machine learning engineers and data scientists in visualizing, analyzing and understanding their data. "Leveraging unstructured and visual data is a significant challenge. Although we've seen recent wins in the transition of capabilities from the lab to production, such as those in ADAS, there remains a difficulty in bringing computer vision capabilities into production," Corso told TechCrunch in an email interview.

Were you unable to attend Transform 2022? Check out all of the summit sessions in our on-demand library now! Computer vision AI models rely on having properly labeled data in order to infer the correct object. The challenge of helping to verify that data used for a model is accurate is one that Ann Arbor, Michigan-based startup Voxel51 is aiming to solve with open-source tools and a commercial service called FiftyOne Teams. Ann Arbor is home to the University of Michigan, which is where Voxel51 cofounder and CEO Jason Corso works as a professor, and where he got the idea to build the new company.

ANN ARBOR--Researchers, hospitals, companies, consumers and government agencies are drowning in data that they can't fully capitalize on. Now, a team from the University of Michigan has received $1.6 million from the Defense Advanced Research Projects Agency to help develop a toolkit so that even nondata scientists can use that data to possibly answer questions and ultimately speed up the process of discovery. The Michigan project seeks to develop algorithms that draw on techniques like machine learning for applications such as understanding video. It is one of 24 projects selected from around the country. DARPA intends to combine techniques from the projects into a central repository.

Ballet Des Moines' artist in residence, Amenda Tate Corso, creates paintings based on the dancers' motion and movement, find out how in this video. While rehearsing Delcid wears a cell phone transmitting a Bluetooth signal to a movement controlled robot creating art based on his motion. Gliding across a small stage, the dancer's arm flourished with the ornamentation of the music and punctuated each pass of the platform with a staccato embellishment. But this particular dancer was distinctly different from the men and women relevéing and jeté-ing nearby. Manibus, as this flapjack-sized dancing robot is called, is the creation of local engineer-turned-artist Amenda Tate Corso, Ballet Des Moines' newest artist-in-residence. The Manibus project, which captures a dancer's movements via a motion-sensor app and translates them into a painting, marks Ballet Des Moines' first foray into the emerging national trend of marrying computational technology with dance.