Skip to main content

Noticias

2026

PlainHealth, a platform for clear doctor–patient communication, was presented at HeMAI 2026, held as part of ACM ICMI 2026.

Lourdes Moreno con participantes de la sesion ACM ICMI2026@HeMAI

Our colleague Lourdes Moreno took part in HeMAI 2026 —Multimodal Interaction with Generative AI Health Applications— on 9 October in Naples, Italy. The workshop was held as part of the 28th ACM International Conference on Multimodal Interaction (ICMI 2026).

Lourdes presented the paper and poster “PlainHealth: A Multimodal Service Suite for Cognitively Accessible Healthcare Communication”, co-authored with our colleagues Paloma Martínez and Jesús M. Sánchez-Gómez and published in the conference’s companion proceedings.

PlainHealth aims to make health information easier for patients, family members and caregivers to understand. The proposed suite combines services for transcription, report generation for different recipients, plain-language adaptation, visual explanations and voice support, delivered through interfaces designed with cognitive accessibility in mind.

The work incorporates quality checks, traceability and human review of AI-generated content. The suite is under development and builds on the group’s existing resources and components, including a text simplification system with human review. The poster presented the proposed architecture and an example of clinical information adapted into a plain-language explanation and a visual care guide.

The workshop brought together research on emotion recognition in consultations, depression and stress detection, accessible conversational agents and AI assistants for self-management. Evaluation and trust in generative AI health applications were also discussed in the keynote and throughout the workshop.

Read the paper

Póster de PlainHealth presentado en HeMAI 2026.

 

The HULAT-UC3M group Presents Research on Making Accessibility Visible: Adaptive and Accessible Interfaces for High-Stress Healthcare Communication at ACM womENcourage 2026

The HULAT-UC3M group Presents Research on Making Accessibility Visible: Adaptive and Accessible Interfaces for High-Stress Healthcare Communication at womENcourage 2026

HULAT at Interacción 2026

Foto de Lourdes Moreno, Raquel Hervás, Virginia Francisco Two talks at XXVI edición del Congreso Internacional de Persona-Ordenador (INTERACCIÓN 2026), held in Albacete 1-4 september.

Article "Benchmark of Small Language Models for Plain-Language Simplification in Spanish Clinical Texts"

resumen gráfico The article "Benchmark of Small Language Models for Plain-Language Simplification in Spanish Clinical Texts" has been published in Scientific Reports journal,

HULAT @ CORIA-TALN 2026

Rémi Cardon presentando en un auditorio. Rémi Cardon represented HULAT-UC3M at CORIA-TALN 2026, presenting three papers across the conference and its affiliated workshops.

HULAT presents its advances in clear language, Easy-to-Read and human-centred artificial intelligence at the Royal Spanish Academy

Paloma Martínez presentando en la RAE

From the HULAT - Human Language and Accessibility Technologies Group, we presented some of our research projects at the event “Clear and Accessible Language and Artificial Intelligence”, held at the Royal Spanish Academy (Real Academia Española) on 26 and 27 May, as a preparatory meeting for the Third Convention of the Pan-Hispanic Network for Clear and Accessible Language.

During the session devoted to cognitive accessibility, Paloma Martínez Fernández presented several lines of work of the group and highlighted the importance of developing artificial intelligence systems supported by evidence obtained through user studies. This evidence is essential for adapting language models to different profiles and needs in tasks involving text simplification, Easy-to-Read and clear language.

Paloma Martínez also emphasised that the quality of the results cannot depend solely on automation. In this type of application, it is essential to integrate expert supervision, both to assess the linguistic and cognitive adequacy of the generated texts and to verify their alignment with standards and with the real needs of the target users.

Within the framework of the HumanAI project (https://humanai.uc3m.es/), we are exploring the generation of clear language and support for Easy-to-Read through openly available language models, with particular attention to small models that can contribute to the development of more energy-sustainable solutions. However, our approach does not propose uncontrolled automation. We believe that the evaluation of these models’ performance should not be delegated solely to automatic metrics, as these do not always capture essential aspects such as text comprehension, meaning preservation or compliance with accessibility and clear language standards.

It is not enough for a model to produce a text that appears to be simpler. It is necessary to check whether it preserves the meaning, avoids removing relevant information, does not add invented information, respects sensitive terms, integrates into a cognitively accessible interface and fits the profile of the target user. There is no universal simplification strategy that is valid for all contexts and users.

From HULAT-UC3M, through the HumanAI project, we advocate for artificial intelligence that does not operate as a black box, but rather as part of a controlled, assessable and governed process, with different levels of human participation. The combination of language models, user-based evidence, expert supervision and standards-based criteria is key to advancing towards more accessible, responsible and human-centred language technologies.

The presentation is available via the link shared by the organisation, from 1:05:
https://www.youtube.com/live/1n8HzzkNAB8?si=ZU5eTrAnRX9ckIVl 

Slide

Voice assistants to improve accessibility

Logotipo de HULAT y CRE HULAT has collaborated with the Spanish Red Cross in a study on the use of voice assistants as allies for accessibility at home. The study focuses on analysing how voice assistants and accessible home automation can support the autonomy, accessibility and daily well-being of people with disabilities and reduced mobility.

Participation en IUI 2026. Human Oversight-by-Design for Accessible Generative IUIs

The HULAT-UC3M research group participates in the AI CHAOS!@IUI 2026 (ACM Conference on Intelligent User Interfaces) with the paper "Human Oversight-by-Design for Accessible Generative IUIs", authored by predoctoral researcher Blessing Jerry, along with Lourdes Moreno and Paloma Martínez.

The work lies at the intersection of Human–Computer Interaction, accessibility, and artificial intelligence, and addresses how to generate accessible user interfaces using language models, aligned with standards and adapted to different user profiles, capabilities, and usage contexts. The proposal is based on the idea that, in sensitive domains, it is not enough to automatically generate content: it is also necessary to ensure that the resulting interface is understandable, usable, and appropriate for diverse users.

The paper proposes an architecture for accessible interface generation based on structured user profiles, adaptation rules, validated templates, and traceability with respect to standards such as WCAG 2.2, EN 301 549, ISO 24495-1, and W3C COGA. Based on these elements, the system can produce adapted interfaces and content, for example, through plain language, step-by-step structuring, pictograms, or visual adjustments tailored to user needs.

In addition, the work incorporates an oversight-by-design approach, where automated evaluation acts as a risk signal and triggers human review and system adjustments when necessary. In this way, accessible interface generation is framed not only as a technical problem, but also as a matter of interaction design, governance, and responsibility in AI systems.

This work is part of the HumanAI project and reinforces the research lines of the HULAT-UC3M group on the design and evaluation of accessible technologies, with a particular focus on human-centered intelligent interfaces.

Preprint available at:
https://arxiv.org/pdf/2602.13745

 

Previous related work: https://arxiv.org/abs/2601.06616 

 

The HULAT-UC3M research group participates in LREC 2026 with a paper on accessible text generation with human participation and supervision

Overall workflow of the proposed HiTL–HoTL framework

The HULAT-UC3M research group will participate in LREC 2026 with the paper “A Human-in/on-the-Loop Framework for Accessible Text Generation”, authored by Lourdes Moreno and Paloma Martínez, and accepted in the track “Simplification, Plain Language and Assistive Technologies”.

The paper proposes a hybrid framework for plain language text generation based on a Human-in/on-the-Loop approach, combining language models with human involvement at different stages of the process. This approach integrates both direct participation during content generation and subsequent supervision of the outputs.

The proposal is structured around the use of cognitive accessibility standards, signal-based triggers for expert review, and quality indicators, with the aim of building a traceable, explainable, and auditable process. In this way, the evaluation of generated texts does not rely solely on automatic metrics, but also incorporates criteria focused on real comprehension and human validation.

This work is part of the HumanAI project and aligns with the research lines of the HULAT-UC3M group at the intersection of Natural Language Processing and Accessibility, with a particular focus on developing AI-based solutions to improve information comprehension for diverse user profiles.

Preprint available at:
https://arxiv.org/pdf/2603.18879

 

Workshop READIxTSAR Joint Workshop on Readability and Text Simplification @ LREC 2026

Rémi Cardon is taking part in the organization of READIxTSAR workshop that will be held in Palma de Mallorca, 11th May 2026 collocated with LREC 2026.