Gonzalo Gabriel Méndez
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Interests
Gonzalo Gabriel Méndez has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Oscar Moreno
Escuela Superior Politécnica del Litoral, Ecuador
4 shared publications - Patricio Mendoza
Escuela Superior Politécnica del Litoral, Ecuador
3 shared publications - Luis Galárraga
Inria, Rennes, Brittany, France, University of Rennes, Rennes, Brittany, France, CNRS, Rennes, Brittany, France, and IRISA, Rennes, Brittany, France
2 shared publications - Marisol Wong-Villacres
Facultad de Ingeniería en Electricidad y Computación, Escuela Superior Politécnica del Litoral, Guayaquil, Ecuador
2 shared publications - Daniel Ochoa
Facultad de Ingeniería en Electricidad y Computación (FIEC) Escuela Superior Politécnica del Litoral (ESPOL) Guayaquil Ecuador
1 shared publication - Doménica Barreiro
Escuela Superior Politécnica del Litoral, Guayaquil, Ecuador
1 shared publication
Latest Publications
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 8801-8812
Whose Name Comes Up? II: Benchmarking and Intervention-Based Auditing of LLM-Based Scholar Recommendation
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 8801-8812
Open AccessWhose Name Comes Up? II: Benchmarking and Intervention-Based Auditing of LLM-Based Scholar Recommendation
- Open Access
- 27 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 52-57
The Limits of Technological Disruption: AI Imaginaries for Parkinson's Care in Low-Resourced Public Health Systems
- 27 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 52-57
Open AccessThe Limits of Technological Disruption: AI Imaginaries for Parkinson's Care in Low-Resourced Public Health Systems
- Open Access
- 13 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-26
Scrollytelling as an Alternative Format for Privacy Policies
Privacy policies are long, complex, and rarely read, which limits their effectiveness in informed consent. We investigate scrollytelling, a scroll-driven narrative approach, as a privacy policy presentation format. We built a prototype that interleaves the full policy text with animated visuals to create a dynamic reading experience. In an online study (N = 454), we compared our tool against text, two nutrition-label variants, and a standalone interactive visualization. Scrollytelling improved user experience over text, yielding higher engagement, lower cognitive load, greater willingness to adopt the format, and increased perceived clarity. It also matched other formats on comprehension accuracy and confidence, with only one nutrition-label variant performing slightly better. Changes in perceived understanding, transparency, and trust were small and statistically inconclusive. These findings suggest that scrollytelling can preserve comprehension while enhancing the experience of policy reading. We discuss design implications for accessible policy communication and identify directions for increasing transparency and user trust.
- 13 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-26
Open AccessScrollytelling as an Alternative Format for Privacy Policies
Privacy policies are long, complex, and rarely read, which limits their effectiveness in informed consent. We investigate scrollytelling, a scroll-driven narrative approach, as a privacy policy presentation format. We built a prototype that interleaves the full policy text with animated visuals to create a dynamic reading experience. In an online study (N = 454), we compared our tool against text, two nutrition-label variants, and a standalone interactive visualization. Scrollytelling improved user experience over text, yielding higher engagement, lower cognitive load, greater willingness to adopt the format, and increased perceived clarity. It also matched other formats on comprehension accuracy and confidence, with only one nutrition-label variant performing slightly better. Changes in perceived understanding, transparency, and trust were small and statistically inconclusive. These findings suggest that scrollytelling can preserve comprehension while enhancing the experience of policy reading. We discuss design implications for accessible policy communication and identify directions for increasing transparency and user trust.
- Open Access
Whose Name Comes Up? Benchmarking and Intervention-Based Auditing of LLM-Based Scholar Recommendation
Large language models (LLMs) are increasingly used for academic expert recommendation. Existing audits typically evaluate model outputs in isolation, largely ignoring end-user inference-time interventions. As a result, it remains unclear whether failures such as refusals, hallucinations, and uneven coverage stem from model choice or deployment decisions. We introduce LLMScholarBench, a benchmark for auditing LLM-based scholar recommendation that jointly evaluates model infrastructure and end-user interventions across multiple tasks. LLMScholarBench measures both technical quality and social representation using nine metrics. We instantiate the benchmark in physics expert recommendation and audit 22 LLMs under temperature variation, representation-constrained prompting, and retrieval-augmented generation (RAG) via web search. Our results show that end-user interventions do not yield uniform improvements but instead redistribute error across dimensions. Higher temperature degrades validity, consistency, and factuality. Representation-constrained prompting improves diversity at the expense of factuality, while RAG primarily improves technical quality while reducing diversity and parity. Overall, end-user interventions reshape trade-offs rather than providing a general fix. We release code and data that can be adapted to other disciplines by replacing domain-specific ground truth and metrics.
Open AccessWhose Name Comes Up? Benchmarking and Intervention-Based Auditing of LLM-Based Scholar Recommendation
Large language models (LLMs) are increasingly used for academic expert recommendation. Existing audits typically evaluate model outputs in isolation, largely ignoring end-user inference-time interventions. As a result, it remains unclear whether failures such as refusals, hallucinations, and uneven coverage stem from model choice or deployment decisions. We introduce LLMScholarBench, a benchmark for auditing LLM-based scholar recommendation that jointly evaluates model infrastructure and end-user interventions across multiple tasks. LLMScholarBench measures both technical quality and social representation using nine metrics. We instantiate the benchmark in physics expert recommendation and audit 22 LLMs under temperature variation, representation-constrained prompting, and retrieval-augmented generation (RAG) via web search. Our results show that end-user interventions do not yield uniform improvements but instead redistribute error across dimensions. Higher temperature degrades validity, consistency, and factuality. Representation-constrained prompting improves diversity at the expense of factuality, while RAG primarily improves technical quality while reducing diversity and parity. Overall, end-user interventions reshape trade-offs rather than providing a general fix. We release code and data that can be adapted to other disciplines by replacing domain-specific ground truth and metrics.
- Open Access
- 16 October 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Human-Computer Interaction
- Vol. 9 (7) , 1-30
The Hidden Workload: Student Data Work in Multimodal Algorithmic Evaluations
As algorithmic systems increasingly mediate human activities across diverse domains, they shift more responsibility for data collection onto users, fundamentally altering the nature of data work. This paper examines the implications of this shift by investigating student-led data collection and automated feedback interpretation using a mobile, multimodal learning analytics (MMLA) tool designed to coach oral presentation skills. Our findings reveal that while this user-controlled data collection provides greater flexibility, it also imposes speculative labor, compelling students to adjust their behavior to align with perceived standards of good data even when such changes are unwarranted. The study highlights the often-overlooked informal labor involved in managing the socio-material conditions of data collection, emphasizing the need for MMLA tools that offer adaptive support and guidance. These insights extend to algorithmic system design in educational and professional contexts, advocating for systems that balance user autonomy with workload-minimizing guidance to achieve equitable accountability.
- 16 October 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Human-Computer Interaction
- Vol. 9 (7) , 1-30
Open AccessThe Hidden Workload: Student Data Work in Multimodal Algorithmic Evaluations
As algorithmic systems increasingly mediate human activities across diverse domains, they shift more responsibility for data collection onto users, fundamentally altering the nature of data work. This paper examines the implications of this shift by investigating student-led data collection and automated feedback interpretation using a mobile, multimodal learning analytics (MMLA) tool designed to coach oral presentation skills. Our findings reveal that while this user-controlled data collection provides greater flexibility, it also imposes speculative labor, compelling students to adjust their behavior to align with perceived standards of good data even when such changes are unwarranted. The study highlights the often-overlooked informal labor involved in managing the socio-material conditions of data collection, emphasizing the need for MMLA tools that offer adaptive support and guidance. These insights extend to algorithmic system design in educational and professional contexts, advocating for systems that balance user autonomy with workload-minimizing guidance to achieve equitable accountability.
- Open Access
- 27 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
The Potential of Cognitive Circles to Measure Mental Load
- 27 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessThe Potential of Cognitive Circles to Measure Mental Load
- Open Access
- 11 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-2
A Tool for Visualizing Flood Impact on Urban Mobility
- 11 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Open AccessA Tool for Visualizing Flood Impact on Urban Mobility
- Open Access
- 11 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Mentorship Navigator: Visual Exploration of Academic Lineages
- 11 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Open AccessMentorship Navigator: Visual Exploration of Academic Lineages
- Open Access
- 22 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-8
The Landscape of Visual Information Communication and Interaction Research
A large body of work has sought to create effective ways of communicating and interacting with visual information. This paper presents insights into this research landscape based on the proceedings from the International Symposium on Visual Information Communication and Interaction (VINCI). We analyze several aspects of VINCI’s scholarship, including key contributors, emerging topics, and collaboration and citation patterns. Our findings reveal the dynamic character of VINCI and its advancements, highlighting its interdisciplinary nature with contributions from diverse institutions and countries, with the majority of authors engaging in both domestic and international collaborations. The insights from this study can inform future research directions and support more fruitful collaborations among VINCI researchers.
- 22 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessThe Landscape of Visual Information Communication and Interaction Research
A large body of work has sought to create effective ways of communicating and interacting with visual information. This paper presents insights into this research landscape based on the proceedings from the International Symposium on Visual Information Communication and Interaction (VINCI). We analyze several aspects of VINCI’s scholarship, including key contributors, emerging topics, and collaboration and citation patterns. Our findings reveal the dynamic character of VINCI and its advancements, highlighting its interdisciplinary nature with contributions from diverse institutions and countries, with the majority of authors engaging in both domestic and international collaborations. The insights from this study can inform future research directions and support more fruitful collaborations among VINCI researchers.
- Open Access
- 22 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-2
An Interactive Visualization Tool for Exploring Implicit Relationships in Relational Datasets
Relational datasets capture insights into complex networks of entities and their interconnections. The analysis of such datasets is critical for various domains, from social and biological networks to scientific research. The complexity and interdependencies inherent to relational datasets present significant challenges for analysts aiming to explore and understand such data. These challenges are particularly notable for individuals lacking expertise in data visualization tools and techniques, as well as those without training in deriving complex relations between the entities contained within the dataset. In this paper, we present a visualization tool designed to facilitate this kind of exploration. We illustrate the use of our tool with a dataset on the scientific production of the VINCI symposium.
- 22 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Open AccessAn Interactive Visualization Tool for Exploring Implicit Relationships in Relational Datasets
Relational datasets capture insights into complex networks of entities and their interconnections. The analysis of such datasets is critical for various domains, from social and biological networks to scientific research. The complexity and interdependencies inherent to relational datasets present significant challenges for analysts aiming to explore and understand such data. These challenges are particularly notable for individuals lacking expertise in data visualization tools and techniques, as well as those without training in deriving complex relations between the entities contained within the dataset. In this paper, we present a visualization tool designed to facilitate this kind of exploration. We illustrate the use of our tool with a dataset on the scientific production of the VINCI symposium.