CALL FOR PAPERS

CFP for the 2nd Workshop on Information Retrieval for Understudied Users (IR4U2) - Bridging User-centered AI with IR: Making Information Retrieval Accessible for All

​

Information Retrieval (IR) remains an active, fast-paced area of research. Nevertheless, most advancements in this domain have primarily favoured what we might term 'conventional' users, such as adults from the global north.

We envision IR4U2 as a forum to spotlight research and industry efforts that, while sparse, consider diverse and often understudied user groups.

Among these groups, we find, for example, older adults, young children and adolescents, individuals afflicted by mental health disorders, users with autism spectrum disorder (ASD), individuals with intellectual disabilities, or those with specific learning needs.

In this second iteration of IR4U2, we will focus more specifically on a **user-centred AI perspective**, which we believe is vital for informing the design, development, and assessment of IR systems that thoughtfully address the diverse needs of understudied populations, ensuring genuine accessibility and inclusivity.

We aim to

• Build, extend and maintain a community involving multi-disciplinary Ph.D. students, academics, and industry practitioners aiming to advance understanding of core (AI for) IR technology and its impact on a broad range of understudied users.

• Take inventory of the different kinds of use cases, research work, ongoing efforts, and existing resources while reflecting on the progress made since the last IR4U2 iteration.

• Define guidelines to promote user-centric design of IR systems.

• Outline a general approach to facilitate research endeavors driven by and at the service of specific user groups

IR4U2 will be a highly participatory, full-day, in-person workshop. This will involve informal interactions, facilitated group work, and brief presentations of accepted contributions. These contributions will help construct a snapshot of the current works in this area.

​

Topics of interest include:

• User-centered design of IR algorithms for understudied users

• User-centric design of user interfaces for IR for understudied users

• The use of AI for enhancing IR systems for understudied users

• The positive and negative impact of AI on IR for understudied users

• User modelling to enable IR and recommendation technologies tailored to understudied populations

• Data collection and benchmark development of IR catering to understudied populations.

• IR applications targeting understudied populations

• IR techniques in the era of LLM and generative models with and for understudied populations

​

This is not an exhaustive list, as we are also interested in contributions discussing challenges inherent to designing IR technology at the service of understudied user groups–from the need for multidisciplinary, multi-stakeholder collaborations to how to build datasets.

Submissions

Peer-reviewed contributions. These include research papers presenting empirical explorations, user studies, and/or new algorithms related to the open problems listed in topics of interest, as well as position/vision contributions discussing uses cases, challenges, and future research directions. These contributions should be submitted on EasyChair in at least 4 pages, but up to 12 (excluding references). Papers should be formatted in CEURART’s single-column template, also available as an Overleaf template.

Accepted contributions will be presented at the workshop and published in workshop proceedings.

Editorially-reviewed contributions.  To enable potential attendees to articulate their views on the workshop topics, share already-published works, bring awareness to ongoing European projects in this area, etc., we ask to submit a short extended abstract. This type of informal submission (via this form: https://bit.ly/ir4u2_2) will be editorially reviewed by workshop organizers to gauge fit as well as themes of interest for group discussions. Accepted contributions will be presented at the workshop, but not published in the proceedings.

​

At least one author of each accepted paper must register and attend the workshop.

Please feel free to reach out to the ir4u2.research@gmail.com with questions.