We’re in the process of curating a list of this year’s publications. If you would like your paper included, please submit it via our dedicated form.
Disclaimer: This list is not complete yet; the DOIs might not be working yet.
Investigating Implicit Adaptivity of an LLM-based Companion for low-acuity Patients in the Emergency Department
Jacobe Klein (Freie Universität Berlin)
Abstract | Tags: Student Mentoring Program (SMP) | Links:
@inproceedings{Klein2026InvestigatingImplicit,
title = {Investigating Implicit Adaptivity of an LLM-based Companion for low-acuity Patients in the Emergency Department},
author = {Jacobe Klein (Freie Universität Berlin)},
url = {https://www.mi.fu-berlin.de/en/inf/groups/hcc/index.html, website
https://www.linkedin.com/in/jacobe-klein, author's linkedin},
doi = {10.1145/3772363.3799203},
year = {2026},
date = {2026-04-13},
urldate = {2026-04-13},
abstract = {In emergency departments (ED), patients' experiences often involve needs beyond medical care, including empathetic communication while feeling seen and informed. My thesis aims to address a care gap caused by staffing constraints and high patient numbers: advancements in artificial intelligence (AI) offer an opportunity to support low-acuity patients during passive waiting times through an AI companion that prepares for consultation, provides information on ED processes, and offers empathetic support without replacing clinical staff. Based on workshops and patient interviews, preliminary requirements for this companion concept were gathered. Building on these insights, my next steps focus on investigating how the companion can implicitly adapt its role and behavior to individual patients, without the cognitive burden of explicitly customizing it. My thesis contributes to the growing body of literature on patient-facing conversational AI and explores how adaptive systems can enhance patient experiences while supporting clinical workflows in high-stress ED settings.},
keywords = {Student Mentoring Program (SMP)},
pubstate = {published},
tppubtype = {inproceedings}
}
In emergency departments (ED), patients' experiences often involve needs beyond medical care, including empathetic communication while feeling seen and informed. My thesis aims to address a care gap caused by staffing constraints and high patient numbers: advancements in artificial intelligence (AI) offer an opportunity to support low-acuity patients during passive waiting times through an AI companion that prepares for consultation, provides information on ED processes, and offers empathetic support without replacing clinical staff. Based on workshops and patient interviews, preliminary requirements for this companion concept were gathered. Building on these insights, my next steps focus on investigating how the companion can implicitly adapt its role and behavior to individual patients, without the cognitive burden of explicitly customizing it. My thesis contributes to the growing body of literature on patient-facing conversational AI and explores how adaptive systems can enhance patient experiences while supporting clinical workflows in high-stress ED settings.