Stella Heras

AAMAS Conference 2026 Conference Paper

Decentralized Value Systems Agreements

• Arturo Hernández-Sánchez
• Natalia Criado
• Stella Heras
• Miguel Rebollo
• Jose Such

One of the biggest challenges of value-based decision-making is dealing with the subjective nature of values. The relative importance of a value for a particular decision varies between individuals, and people may also have different interpretations of what aligning with a value means in a given situation. While members of a society are likely to share a set of principles or values, their value systems—that is, how they interpret these values and the relative importance they give to them—have been found to differ significantly. This work proposes a novel method for aggregating value systems, generating distinct value agreements that accommodate the inherent differences within these systems. Unlike existing work, which focuses on finding a single value agreement, the proposed approach may be more suitable for a realistic and heterogeneous society. In our solution, the agents indicate their value systems and the extent to which they are willing to concede. Then, a set of agreements is found, taking a decentralized optimization approach. Our work has been applied to identify value agreements in two real-world scenarios using data from a Participatory Value Evaluation process and a European Value Survey. These case studies illustrate the different aggregations that can be obtained with our method and compare them with those obtained using existing value system aggregation techniques. In both cases, the results showed a substantialimprovementinindividualutilitiescomparedtoexisting alternatives.

AAMAS Conference 2026 Conference Paper

Exploring Cognitive Bias Impact, Detection and Mitigation in Large Language Models

• Ana Gutiérrez-Mandingorra
• Stella Heras
• Javier Palanca
• Vicent Botti

Large Language Models have revolutionized a wide range of domains—including education, healthcare, law, and industry—by enabling the automation of complex tasks through advanced natural language understanding and text generation. However, theirwidespreaddeploymenthasraisedsignificantethical and practical concerns, particularly regarding the biases embedded in their outputs. While social biases in LLMs have been extensively examined across the literature, cognitive biases—systematic patterns of deviation from normative reasoning rooted in human cognition—remain comparatively underexplored. These biases pose a unique challenge, as they can be subtly introduced, for instance, through prompt design or inherited from training data. Therefore, the study of cognitive biases in LLMs represents an emerging and increasingly critical area of research. This work presents a structured investigation into the presence, detection, and mitigation of cognitive biases in LLMs. We propose a three-stage experimental strategy: (1) evaluating the influence of prompt-induced cognitive biases on model outputs, (2) exploring bias detection strategies based on Retrieval-Augmented Generation systems enhanced with cognitive theory knowledge and incorporating agent-based reasoning elements, and (3) mitigating bias effects through warning-based interventions. Our findings aim to contribute towards a better understanding of LLMs’ alignment with human cognition and offer a foundation for safer and more trustworthy AI systems.

AAMAS Conference 2026 Conference Paper

SofIA: AI Clinical Companion for Real-Time Documentation and Decision Support

• Leire Villarroya-Martínez
• Enrique Alcázar Garzás
• Stella Heras
• Javier Palanca
• Vicent Botti

We present SofIA, a hospital-ready assistant that helps doctors with three everyday tasks: (1) summarising a patient’s history, (2) answeringclinicallookupsfromhospitalguidelines, and(3)drafting structured clinical notes during or after the visit. The demo focuses on the user experience rather than algorithms. Users can try SofIA on a set of synthetic patients and see how summaries, answers with cited sources, and draft notes are produced in seconds. We will also show how SofIA connects to hospital systems and how safety checks keep humans in control.

AAMAS Conference 2026 Conference Paper

TEME: A Multi-Agent Evaluation Framework for Spanish Medical Speech Recognition

• Leire Villarroya-Martinez
• Stella Heras
• Javier Palanca
• Vicent Botti
• Edwin Tadeo-Gomez
• Enrique Alcazar Garzas

Conventional metrics like Word Error Rate (WER) fail to differentiate between minor variations and potentially life-threatening medical mistakes. We introduce TEME (Medical Accuracy Test in Spanish), a supervised multi-agent evaluation framework for Spanish medical Automatic Speech Recognition (ASR). TEME employs a two-layer architecture with specialized agents assessing transcriptions for clinical awareness, overseen by a consensus agent that applies safety rules. Testing on 90 validated clinical dialogues shows that TEME successfully captures clinically relevant error severity that conventional metrics miss, providing a safety-aware alternative for medical ASR evaluation.

JAAMAS Journal 2025 Journal Article

An introduction to computational argumentation research from a human argumentation perspective

• Ramon Ruiz-Dolz
• Stella Heras
• Ana García-Fornes

Abstract Computational Argumentation studies how human argumentative reasoning can be approached from a computational viewpoint. Human argumentation is a complex process that has been studied from different perspectives (e. g. , philosophical or linguistic) and that involves many different aspects beyond pure reasoning, such as the role of emotions, values, social contexts, and practical constraints, which are often overlooked in computational approaches to argumentation. The heterogeneity of human argumentation is present in Computational Argumentation research, in the form of various tasks that approach the main phases of argumentation individually. With the increasing interest of researchers in Artificial Intelligence, we consider that it is of great importance to provide guidance on the Computational Argumentation research area. Thus, in this paper, we present a general overview of Computational Argumentation, from the perspective of how humans argue. For that purpose, the following contributions are produced: (i) a consistent structure for Computational Argumentation research mapped with the human argumentation process; (ii) a collective understanding of the tasks approached by Computational Argumentation and their synergies; (iii) a thorough review of important advances in each of these tasks; and (iv) an analysis and a classification of the future trends in Computational Argumentation research and relevant open challenges in the area.

ECAI Conference 2024 Conference Paper

A Predictive Model for Risk Management of the Hospitalised Patient as a Clinical Decision Support System

• Isabel Benlloch-Blasco
• Vicente J. Botti
• Stella Heras
• M. Isabel Marmol
• Isabel Miguel
• Javier Palanca
• Antonio Ruiz

Assessing patient risk or degree of vulnerability is crucial for effective nursing care, resource allocation and prevention of complications and adverse events. A scale-based system evaluates hospitalised patients, capturing their status across domains by aggregating and weighting scale scores. Currently, in most hospitals, nurses manually interpret these scores and record the data. Even in high-tech hospitals with electronic health records (EHRs) the interpretation of results still depends on manual analysis, leading to subjectivity and errors. The aim of this project is to develop a predictive system that is seamlessly integrated into the EHR as a decision support tool. The system will categorise patients based on their need for resource allocation, which is interpreted as risk. Real data from 1800 patients were analysed and showed 86% of accuracy in predicting risk using a decision tree algorithm.

AAMAS Conference 2018 Conference Paper

An Argumentation-based Conversational Recommender System for Recommending Learning Objects

• Javier Palanca
• Stella Heras
• Paula Rodr�guez Mar�n
• N�stor Duque
• Vicente Juli�n

In this paper, we present an argumentation-based Conversational Educational Recommender System (C-ERS), which helps students to find the more suitable learning resources considering their learning objectives and profile. The recommendation process is based on an argumentation-based technique, which selects those learning objects (LOs) for which it is able to generate a greater number of arguments justifying their suitability.

EUMAS Conference 2016 Conference Paper

Using Argumentation Schemes for a Persuasive Cognitive Assistant System

• Ângelo Costa
• Stella Heras
• Javier Palanca
• Jaume Jordán
• Paulo Novais
• Vicente Julián

Abstract The iGenda framework is a cognitive assistant that helps care-receivers and caregivers in the management of their agendas. One of the problems detected in systems of this kind is the lack of user engagement. This engagement can be improved through the application of persuasion techniques in order to convince users to act in a specific way. According to this, this paper presents a new architecture that will allow the system to select and recommend activities that potentially best suits to the users’ interests based on argumentation techniques.

EUMAS Conference 2015 Conference Paper

Argumentation-Based Hybrid Recommender System for Recommending Learning Objects

• Paula Rodríguez
• Stella Heras
• Javier Palanca
• Néstor D. Duque
• Vicente Julián

Abstract Recommender Systems aim to provide users with search results close to their needs, making predictions of their preferences. In virtual learning environments, Educational Recommender Systems deliver learning objects according to the student’s characteristics, preferences and learning needs. A learning object is an educational content unit, which once found and retrieved may assist students in their learning process. In previous work, authors have designed and evaluated several recommendation techniques for delivering the most appropriate learning object for each specific student. Also, they have combined these techniques by using hybridization methods, improving the performance of isolated techniques. However, traditional hybidization methods fail when the learning objects delivered by each recommendation technique are very different from those selected by the other techniques (there is no agreement about the best learning object to recommend). In this paper, we present a hybrid recommendation method based on argumentation theory that combines content-based, collaborative and knowledge-based recommendation techniques and provides the students with those objects for which the system is able to generate more arguments to justify their suitability. This method has been tested by using a database with real data about students and learning objects, getting promising results.

EAAI Journal 2014 Journal Article

Modelling dialogues in agent societies

• Stella Heras
• Vicente Botti
• Vicente Julián

Besides the simpler ability to interact, open multi-agent systems must include mechanisms for their agents to reach agreements by taking into account their social context. Argumentation provides multi-agent systems with a framework that assures a rational communication, which allows agents to reach agreements when conflicts of opinion arise. In this paper, we present the dialogue protocol that agents of a case-based argumentation framework can use to interact when they engage in argumentation dialogues. The syntax and semantics of the argumentation protocol are formalised and discussed. To illustrate our proposal, we have applied the protocol in the context of a water market. By using our dialogue protocol, agents represent water users that are able to explore different water allocations and justify their views about what is the best water distribution in a certain environment.

AAMAS Conference 2010 Conference Paper

Strategic Argumentation in Open Multi-Agent Societies

• Stella Heras

KER Journal 2009 Journal Article

Challenges for a CBR framework for argumentation in open MAS

• Stella Heras
• Vicente Botti
• Vicente Julián

Abstract Nowadays, Multi-Agent Systems (MAS) are broadening their applications to open environments, where heterogeneous agents could enter into the system, form agents’ organizations and interact. The high dynamism of open MAS gives rise to potential conflicts between agents and thus, to a need for a mechanism to reach agreements. Argumentation is a natural way of harmonizing conflicts of opinion that has been applied to many disciplines, such as Case-Based Reasoning (CBR) and MAS. Some approaches that apply CBR to manage argumentation in MAS have been proposed in the literature. These improve agents’ argumentation skills by allowing them to reason and learn from experiences. In this paper, we have reviewed these approaches and identified the current contributions of the CBR methodology in this area. As a result of this work, we have proposed several open issues that must be taken into consideration to develop a CBR framework that provides the agents of an open MAS with arguing and learning capabilities.

ECAI Conference 2006 Conference Paper

CBR-TM: A New Case-Based Reasoning System for Help-Desk Environments

• Juan Ángel García-Pardo
• Stella Heras
• Rafael Ramos-Garijo
• Alberto Palomares
• Vicente Julián
• Miguel Rebollo
• Vicente J. Botti

In this paper, a new CBR system for help-desk environments is presented. This CBR system provides intelligent customer support for multiple domains. It is also portable and flexible. The system is implemented as a module of a complete help-desk application to make it as independent as possible of any change in the help-desk system. Each phase of the reasoning cycle is also separated as an independent module, making the CBR system easy to update. The system has been tested in a real call center managed by the Spanish company TISSAT S. A.