Towards Trustworthy Multimodal AI Systems
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (47) · 39807-39808
DOI: 10.1609/aaai.v40i47.41338 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Chirag Agarwal.
Abstract (Originalsprache)
Machine learning models have become ubiquitous in the last decade, and with their increasing use in critical applications (e.g., healthcare, financial systems, and crime forecasting), it is vital to ensure that ML developers and practitioners understand and trust their decisions. This problem has become paramount in the era of frontier models, which are developed by training billion-parameter models on broad, uncurated datasets and extensive computing. In this talk, we will first explore the (un)reliability of existing multimodal explainability techniques in large language and multimodal models and understand the robustness and safety implications of Mechanistic Interpretability tools. Next, we will delve into two complementary threads: i) domain-specific safety and related trustworthy evaluation that surfaces risks missed by generic red-teaming, focusing on multilingual and distribution-shifted settings; and ii) methods that explicitly train and assess reasoning in medical LLMs.
Zitieren
APA
Agarwal, C. (2026). Towards Trustworthy Multimodal AI Systems. Proceedings of the AAAI Conference on Artificial Intelligence, 40(47), 39807-39808. https://doi.org/10.1609/aaai.v40i47.41338
Vancouver
Agarwal C. Towards Trustworthy Multimodal AI Systems. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(47):39807-39808. doi:10.1609/aaai.v40i47.41338
BibTeX
@article{Agarwal2026,
author = {Agarwal, Chirag},
title = {Towards Trustworthy Multimodal AI Systems},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {47},
pages = {39807--39808},
doi = {10.1609/aaai.v40i47.41338}
}
EUVIMED