M.A. (Matias) Valdenegro Toro, Dr

Assistant Professor for Machine Learning

Faculteit Science and Engineering

Telefoon:

050 36 36533

E-mail:

m.a.valdenegro.toro rug.nl

ORCID Research Portal

Research interests

My research interests are in building trustworthy machine learning models with real-world data for several applications, including medical AI and Robotics. These relate to Uncertainty Estimation, Computer Vision, and Explainable AI.

Publicaties

Adaptive Prompt Tuning: Vision Guided Prompt Tuning with Cross-Attention for Fine-Grained Few-Shot Learning

Brouwer, E., Woerden, J. E. V., Burghouts, G., Valdenegro-Toro, M. & Zullich, M., 2025, Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. Vol. 2. blz. 114-125 12 blz. (Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications).

Onderzoeksoutput › › peer review

Can Bayesian Neural Networks Explicitly Model Input Uncertainty?

Valdenegro-Toro, M. & Zullich, M., 2025, Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3 VISAPP: VISAPP. Bashford-Rogers, T., Meneveaux, D., Ammi, M., Ziat, M., Jänicke, S., Purchase, H., Radeva, P., Furnari, A., Bouatouch, K. & Sousa, A. A. (uitgave). SCITEPRESS – Science and Technology Publications, Vol. 3. blz. 188-199 12 blz.

Onderzoeksoutput › › peer review

Uncertainty Estimation for Super-Resolution Using ESRGAN

Adapa, M. S., Zullich, M. & Valdenegro-Toro, M., 2025, Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. SciTePress, Vol. 2. blz. 367-374 8 blz. (Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications).

Onderzoeksoutput › › peer review

A Neuromorphic Approach to Obstacle Avoidance in Robot Manipulation

Abdelrahman, A. F., Valdenegro-Toro, M., Bennewitz, M. & Plöger, P. G., 8-apr.-2024, (E-pub ahead of print) In: International Journal of Robotics Research. 36 blz.

Onderzoeksoutput › › peer review

How disentangled are your classification uncertainties?

de Jong, I. P., Sburlea, A. I. & Valdenegro-Toro, M., 22-aug.-2024, 11 blz.

Onderzoeksoutput: Voordruk ›

Know What You do Not Know: Verbalized Uncertainty Estimation Robustness on Corrupted Images in Vision-Language Models

Borszukovski, M., de Jong, I. P. & Valdenegro-Toro, M., 4-apr.-2024, (Submitted) TrustNLP Workshop @ NAACL 2025: Fifth Workshop on Trustworthy Natural Language Processing. Association for Computational Linguistics, ACL Anthology

Onderzoeksoutput › › peer review

Mapping Transformer Leveraged Embeddings for Cross-Lingual Document Representation

Tashu, T. M., Kontos, E.-R., Sabatelli, M. & Valdenegro-Toro, M., 12-jan.-2024, (Submitted) arXiv, 12 blz.

Onderzoeksoutput: Voordruk ›

Overconfidence is Key: Verbalized Uncertainty Evaluation in Large Language and Vision-Language Models

Groot, T. & Valdenegro-Toro, M., 5-mei-2024, (Submitted) arXiv, 27 blz.

Onderzoeksoutput: Voordruk ›

Transferring BCI models from calibration to control: Observing shifts in EEG features

de Jong, I. P., van den Wittenboer, L., Valdenegro-Toro, M. & Sburlea, A. I., 9-sep.-2024, Proceedings of the 9th Graz Brain-Computer Interface Conference 2024. Müller-Putz, G., Kostoglou, K., Oberndorfer, M. E. & Wriessnegger, S. C. (uitgave). Verlag der Technischen Universität Graz, blz. 156-161 6 blz.

Onderzoeksoutput › › peer review

Uncertainty Quantification for cross-subject Motor Imagery classification

Manivannan, P., de Jong, I. P., Valdenegro-Toro, M. & Sburlea, A. I., 9-sep.-2024, Proceedings of the 9th Graz Brain-Computer Interface Conference 2024. Verlag der Technischen Universität Graz, blz. 86-91 6 blz.

Onderzoeksoutput › › peer review

Alle (25) publicaties

Pers/media

Participation in Podcast "Hacia Afuera"

Valdenegro Toro, M.

06/01/2023

Pers / media: Activiteiten met een maatschappelijk belang ›

Alle (1) pers/media

Bekijk volledig onderzoeksprofiel