The 30th ACM Conference on Intelligent User Interfaces (ACM IUI) 2025 took place in Cagliari (Italy) from 24th to 27th March. The IUI is a premier venue for research and developments on intelligent user interfaces. With nine contributions, IML was very prominently represented at this year’s conference.
At the workshop sessions, Aliki Anagnostopoulou presented an extended abstract on self-improving contextualized image captioning. Philipp Olschewski presented TextVision, a project developed by a group of students in a course at the Carl von Ossietzky Universität Oldenburg (PG AIM). Additionally, Robert Leist presented his work “An AI-driven Clinical Decision Support System for the Treatment of Diabetic Retinopathy and Age-related Macular Degeneration”. Siting Liang delivered a presentation on Explainable Biomedical Claim Verification with Large Language Models at the Adaptive XAI workshop, followed by an engaging discussion with the organizers and audience.
During the doctoral consortium session, early-stage researchers had the chance to present and discuss their research with expert mentors. Rida Saghir presented her paper, “Flexible and Interpretable Soundscape Analysis for Biodiversity Assessment and Ecosystem Health for Domain Experts“. The research aims to develop methods for soundscape analysis, create impactful visualizations, and ensure interpretability for ecologists through an interactive interface. It explores how to streamline analysis workflows, effectively analyze multivariate PAM data, enhance user understanding with interactive visualizations, and leverage NLP to support ecological decision-making.
At the demo session, Rida Saghir presented Ecoscape Analyzer, a tool developed to help researchers and ecologists explore long-term, unlabeled passive acoustic monitoring datasets. Beyond offering state-of-the-art methods for soundscape analysis, the tool provides a customizable pipeline for rapid prototyping, supporting tasks such as feature extraction, dimensionality reduction, clustering, visualization, and evaluation of selected analysis pipelines. Sara-Jane Bittner gave a presentation about her Master thesis on a human-centred and ML-supported learning tool, which was developed in context of the flagship project Ophthalmo-AI. Attendees were able to test novel advances in intelligent user interfaces ranging from LLM-supported chatbots up to a game steered with brain-computer-interfacing. In the same session, Robert Leist, László Kopácsi, and Sara-Jane Bittner presented a VR photobook tool in which users can co-create photobooks via different modality combinations in human-AI collaboration.
In the “Interactive Machine Learning Session” Robert Leist presented his full paper “Towards Trustable Clinical Decision Support Systems: A User Study with Ophthalmologists”. The aim was to evaluate a clinical decision support system in terms of familiarity, efficiency, informedness and user experience of the doctors who used it. With those numerous contributions, the IUI 2025 was a great success for the IML researchers.
Sara-Jane Bittner, László Kopácsi, and Robert Leist present their VR photobook tool
Robert Leist presents his full paper in Cagliari
