iSchool researchers receive best paper award at TrustNLP

Sullam Jeoung
Sullam Jeoung
Jana Diesner
Jana Diesner, Affiliate Associate Professor
Halil Kilicoglu
Halil Kilicoglu, Associate Professor

A paper coauthored by PhD student Sullam Jeoung, Associate Professor Jana Diesner, and Associate Professor Halil Kilicoglu was named Best Long Paper at TrustNLP: Third Workshop on Trustworthy Natural Language Processing, which was held in conjunction with the Annual Conference of the Association for Computational Linguistics (ACL 2023). In their paper, "Examining the Causal Impact of First Names on Language Models: The Case of Social Commonsense Reasoning," the researchers discuss how the use of first names in language models can impact their trustworthiness.

"Language models are increasingly used and deployed across various applications and domains that engage with users, ranging from some recent popular models such as ChatGPT and Bard to applications such as AI counseling," explained Jeoung. "As language models are used in circumstances where social intelligence and commonsense reasoning are becoming important, it is imperative to ensure the trustworthiness of language models."

Through a controlled experiment to measure the causal effect of first names on reasoning, the researchers were able to distinguish between model predictions due to chance and those caused by factors of interest. Their results indicate that the frequency of first names in a model has a direct effect on its prediction.

"Our findings suggest that to ensure model robustness, it is essential to augment datasets with more diverse first names during the configuration stage," said Jeoung.

Updated on
Backto the news archive

Related News

Carboni joins the iSchool faculty

The iSchool is pleased to announce that Nicola Carboni has joined the faculty as an assistant professor. He previously served as a postdoctoral researcher and lecturer in digital humanities at the University of Geneva.

Nicola Carboni

Youth-AI-Safety named a winning team in international hackathon

A team of researchers from the SALT (Social Computing Systems) Lab has been selected as a winner in an international hackathon hosted by the Berkeley Center for Responsible, Decentralized Intelligence. The LLM Agents MOOC Hackathon brought together over 3,000 students, researchers, and practitioners from 127 countries to build and showcase innovative work in large language model (LLM) agents, grow the AI agent community, and advance LLM agent technology.

Chan to present "Predatory Data" work at named lectures

Associate Professor Anita Say Chan will present research drawn from her new book, Predatory Data: Eugenics in Big Tech and Our Fight for an Independent Future, at two named lectures this month. The lectures, which celebrate Women's History Month, will be held at the University of Minnesota and Carnegie Mellon University.

Anita Say Chan

New home for the Center for Children’s Books

The Center for Children's Books (CCB) at the iSchool is a crossroads for critical inquiry, professional training, and educational outreach related to youth-focused resources, literature, and librarianship. The CCB houses a non-circulating research collection of children’s and young adult books, with emphasis placed on books published within the last two years. The CCB recently moved to a new home in the iSchool building at 501 East Daniel Street. 

inside the Center for Children's Books with colorful furniture and carpet and bookcases.