Hamed Karimi

Hamed Karimi

Machine Learning Researcher (Postdoctoral Fellow) in Computer Engineering.
Working on Uncertainty Quantification in Trustworthy AI and Machine Learning.

Selected Publications

You can also find my research papers on my Google Scholar profile.

Conference Papers

LLMs Uncertainty Quantification via Adaptive Conformal Semantic Entropy

Published in Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), 2026

Prompt-level and response-level uncertainty quantification in LLMs and adaptive accept/abstain decision rule using conformal semantic entropy.

Recommended Citation: Karimi, H., Meyappan, V., & Samavi, R. (2026). “LLMs Uncertainty Quantification via Adaptive Conformal Semantic Entropy.” Accepted to be published in the Proceedings of the 35th International Joint Conference on Artificial Intelligence (IJCAI 2026).
Download Paper

CEAR: Certified Ensemble Adversarial Robustness in DNNs

Published in Proceedings of the Canadian Conference on Artificial Intelligence (Canadian AI), 2026

An ensemble-based robust method that utilizes a hybrid of empirical and certified defense mechanisms.

Recommended Citation: Sadig, D., Maleki, M., Karimi, H., & Samavi, R. (2026). “CEAR: Certified Ensemble Adversarial Robustness in DNNs.” Accepted to be published in the Proceedings of the 39th Canadian Conference on Artificial Intelligence (Canadian AI 2026).
Download Paper

SACP: Spatially-Adaptive Conformal Prediction in Uncertainty Quantification of Medical Image Segmentation

Published in Proceedings of Medical Imaging with Deep Learning (MIDL), PMLR, 2025

Spatially-adaptive conformal prediction for anatomically informed uncertainty sets in medical image segmentation.

Recommended Citation: Bereska, J. I., Karimi, H., & Samavi, R. (2025). “SACP: Spatially-Adaptive Conformal Prediction in Uncertainty Quantification of Medical Image Segmentation.” Accepted in the conference on Medical Imaging with Deep Learning (MIDL 2025). To be published in Proceedings of Machine Learning Research (PMLR).
Download Paper

Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction

Published in Proceedings of the Symposium on Conformal and Probabilistic Prediction with Applications (COPA), PMLR, 2024

Evidential Conformal Prediction (ECP) for adaptive prediction sets with coverage guarantees.

Recommended Citation: Karimi, H., & Samavi, R. (2024). “Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction.” In Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications. Proceedings of Machine Learning Research (PMLR), vol. 230, 466–489.
Download Paper

DENL: Diverse Ensemble and Noisy Logits for Improved Robustness of Neural Networks

Published in Proceedings of the Asian Conference on Machine Learning (ACML), PMLR, 2024

Two-phase training with diversified ensembles and noisy logits for improved adversarial robustness.

Recommended Citation: Yazdani, M., Karimi, H., & Samavi, R. (2024). “DENL: Diverse Ensemble and Noisy Logits for Improved Robustness of Neural Networks.” In Proceedings of the 15th Asian Conference on Machine Learning. Proceedings of Machine Learning Research (PMLR), vol. 222, 1574–1589.
Download Paper

Quantifying Deep Learning Model Uncertainty in Conformal Prediction

Published in Proceedings of the AAAI Symposium Series (Human AI), 2023

Certified quantification of uncertainty derived from conformal prediction sets.

Recommended Citation: Karimi, H., & Samavi, R. (2023). “Quantifying Deep Learning Model Uncertainty in Conformal Prediction.” The Proceedings of the AAAI Symposium Series, 1(1), pp. 142–148.
Download Paper

Ontology Alignment Using Inductive Logic Programming

Published in Proceedings of the International Conference on Web Research (ICWR), IEEE Xplore, 2018

Mapping ontologies using inverse resolution in inductive logic programming.

Recommended Citation: Karimi, H., & Kamandi, A. (2018). “Ontology Alignment Using Inductive Logic Programming.” In 4th International Conference on Web Research (ICWR), 118–127. IEEE.
Download Paper

Journal Articles

A Learning-based Ontology Alignment Approach Using Inductive Logic Programming

Published in Journal of Expert Systems with Applications (ESWA), Elsevier, 2019

Ontology alignment via learning with Inductive Logic Programming (ILP).

Recommended Citation: Karimi, H., & Kamandi, A. (2019). “A Learning-based Ontology Alignment Approach Using Inductive Logic Programming.” Journal of Expert Systems with Applications (ESWA), Elsevier, 125(C), 412–424.
Download Paper

Application-Based Review of Recent Advances of Data Mining in Healthcare

Published in Journal of Biostatistics and Epidemiology (JBE), 2019

Review of data mining advances in healthcare with an application-based perspective.

Recommended Citation: Shirazi, S., Baziyad, H., & Karimi, H. (2019). “An Application-Based Review of Recent Advances of Data Mining in Healthcare.” Journal of Biostatistics and Epidemiology, 5(4).
Download Paper

YAD: A Learning-based Inductive Logic Programming Tool

Published in Journal of Open Source Software (JOSS), 2018

An open-source, Inductive Logic Programming tool for multi-tabular and multi-dimensional learning based on inverse resolution.

Recommended Citation: Kamandi, A., & Karimi, H. (2018). “YAD: A Learning-based Inductive Logic Programming Tool.” Journal of Open Source Software (JOSS), 3(30), 892.
Download Paper