The 2024 AMIA Annual Symposium (San Francisco, California) will be held November 9-11, and the 2024 Conference on Empirical Methods in Natural Language Processing (Miami, Florida) will be held November 12-16. Several University of Michigan School of Information researchers will be presenting their work. 

WORKSHOPS (AMIA)

Research Methods to Advance Health and Healthcare Equity within Health Informatics: An Instructional Workshop

Tiffany Veinot, Marcy Antonio, Alicia Williamson, Lorraine R. Buis, Megan Threats, Ashley Bhogal, Jian-Sin Lee, VG Vinod Vydiswaran

In this workshop attendees will be introduced to a systematic process for integrating health equity into all stages of a project using the “Equitable Research PRAXIS” framework. The PRAXIS framework advocates for attention to four methodological considerations across project stages: Participation and Representation (PR); Appropriate Methods and Interventions (A); ConteXtualization and Structural Competence (X); and Investigation of Systematic Differences (IS). It offers systematic equity considerations for researchers and was derived from nine methodologically diverse equity-focused health informatics studies. The workshop will be divided into didactic instruction and collaborative exercises. Presentations from facilitators with varied expertise will introduce the PRAXIS framework. Breakout sessions will have the attendees collaboratively apply the PRAXIS methodological considerations to case studies throughout four stages of a research study. Attendees will choose from case studies with diverse methods: (1) natural language processing/machine learning; (2) user-centered design; (3) mixed methods; and (4) randomized controlled trials. Discussions will also draw examples from attendees’ experiences with health equity research. The workshop will conclude with a facilitated brainstorming exercise regarding how to advance health equity in attendees’ current or future research. People who are both more and less familiar with health equity research approaches are encouraged to attend.

POSTERS (AMIA)

Training Longformer Models to Infer Surgery Outcomes in Inflammatory Bowel Disease Patients. 

V.G.Vinod Vydiswaran, Deahan Yu, Jiazhao Li, Ryan Stidham

Exploring Challenges and Opportunities in Government-Academic Data Collaborations for Public Health Crisis Response: A Qualitative Study

Jian-Sin Lee, Tiffany Veinot, Elizabeth Yakel

PAPERS (AMIA)

“Getting people access to services is also getting them access to a phone”: Clarifying digital divide dynamics and their consequences in Community Mental Health Care

Alicia Williamson, Tiffany Veinot

PAPERS (EMNLP)

Divide-or-Conquer? Which Part Should You Distill Your LLM?

Zhuofeng Wu, Richard He Bai, Aonan Zhang, Jiatao Gu, V.G.Vinod Vydiswaran, Navdeep Jaitly, Yizhe Zhang 

Investigating the Generalizability of Pretrained Language Models across Multiple Dimensions: A Case Study of NLI and MRC

Ritam Dutt, Sagnik Ray Choudhury, Varun Venkat Rao, Carolyn Rose, V.G.Vinod Vydiswaran

The Language of Trauma: Modeling Traumatic Event Descriptions Across Domains with Explainable AI
Miriam Schirmer, Tobias Leemann, Gjergji Kasneci, Jürgen Pfeffer, David Jurgens

ValueScope: Unveiling Implicit Norms and Values via Return Potential Model of Social Interactions
Chan Young Park, Shuyue Stella Li, Hayoung Jung, Svitlana Volkova, Tanushree Mitra, David Jurgens, Yulia Tsvetkov

Tab2Text - A framework for deep learning with tabular data
Tong Lin*, Jason Yan*, David Jurgens, Sabina Tomkins

Is "A Helpful Assistant" the Best Role for Large Language Models? A Systematic Evaluation of Social Roles in System Prompts
Mingqian Zheng, Jiaxin Pei, Lajanugen Logeswaran, Moontae Lee, David Jurgens

Retrieval augmented generation or long-context llms? a comprehensive study and hybrid approach

Zhuowan Li, Cheng Li, Mingyang Zhang, Qiaozhu Mei, Michael Bendersky

Closing the Loop: Learning to Generate Writing Feedback via Language Model Simulated Student Revisions

Inderjeet Nair, Jiaye Tan, Xiaotian Su, Anne Gere, Xu Wang, Lu Wang

RELATED

See more research by subscribing to UMSI’s free research roundup newsletter today!