Publications
See also my Google Scholar profile.
2026
- International Conference on Learning Representations (ICLR 2026)PDF
@inproceedings{bakman2026iclr, title={Uncertainty as Feature Gaps: Epistemic Uncertainty Quantification of {LLMs} in Contextual Question-Answering}, author={Bakman, Yosef Fridman and Kang, Sungmin and Huang, Zheng and Yaldiz, D. Nezih and Belem, Catarina G. and Zhu, Chen and Kumar, Aakriti and Samuel, Anand and Liu, Dong and Avestimehr, Salman and Karimireddy, Sai Praneeth}, booktitle={International Conference on Learning Representations}, year={2026} } - Under reviewPDF
@article{steyvers2026metacognition, title={Improving Metacognition and Uncertainty Communication in Language Models}, author={Steyvers, Mark and Belem, Catarina G. and Smyth, Padhraic}, year={2026}, note={Under review} } - Under reviewPDF
@article{belem2026certainty, title={From `{May}' to `{Is}': Certainty Distortion in Language Model Rewriting}, author={Belem, Catarina G. and Wu, Steven and Yao, Hao and Steyvers, Mark and Singh, Sameer and Smyth, Padhraic}, year={2026}, note={Under review} }
2025
- NeurIPS 2025 LLM Evaluations WorkshopPDF
@inproceedings{longjohn2025llmevals, title={Bayesian Evaluation of Blackbox {LLM} Behavior}, author={Longjohn, Ryan and Wu, Steven and Kher, Sonali and Belem, Catarina G. and Smyth, Padhraic}, booktitle={NeurIPS Workshop on LLM Evaluations}, year={2025} } - NeurIPS 2025 Workshop on Structured Probabilistic Inference and Generative Modeling (SPIGM)PDF
@inproceedings{ahmed2025spigm, title={Semantic Probabilistic Control of Language Models}, author={Ahmed, Kareem and Belem, Catarina G. and Smyth, Padhraic and Singh, Sameer}, booktitle={NeurIPS Workshop on Structured Probabilistic Inference and Generative Modeling}, year={2025} } - EMNLP 2025 Workshop on Text Simplification, Accessibility, and Readability (TSAR)PDF
@inproceedings{belem2025tsar, title={Readability Reconsidered: A Cross-Dataset Analysis of Reference-Free Metrics}, author={Belem, Catarina G. and Glenn, Paul and Samuel, Anand and Kumar, Aakriti and Liu, Dong}, booktitle={EMNLP Workshop on Text Simplification, Accessibility, and Readability}, year={2025} } - Findings of the Annual Conference of the North American Chapter of the ACL (NAACL 2025)PDF
@inproceedings{belem2025naacl, title={From Single to Multi: How {LLMs} Hallucinate in Multi-Document Summarization}, author={Belem, Catarina G. and Pezeshkpour, Pouya and Iso, Hayate and Maekawa, Seiji and Bhutani, Nikita and Hruschka, Estevam}, booktitle={Findings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics}, year={2025} } - Nature Machine IntelligencePDF
@article{steyvers2025nature, title={What Large Language Models Know and What People Think They Know}, author={Steyvers, Mark and Tejeda, Heliodoro and Kumar, Aakriti and Belem, Catarina G. and Karny, Sherry and Hu, Xinyue and Mayer, Louis W. and Smyth, Padhraic}, journal={Nature Machine Intelligence}, year={2025}, doi={10.1038/s42256-024-00976-7} }
2024
- Empirical Methods in Natural Language Processing (EMNLP 2024)PDF
@inproceedings{belem2024emnlp, title={Perceptions of Linguistic Uncertainty by Language Models and Humans}, author={Belem, Catarina G. and Kelly, Matthew and Singh, Sameer and Steyvers, Mark and Smyth, Padhraic}, booktitle={Proceedings of the Conference on Empirical Methods in Natural Language Processing}, pages={8467--8502}, year={2024} } - TrustNLP Workshop at NAACL 2024 ★ Runner-Up Best Short PaperPDF
@inproceedings{belem2024trustnlp, title={Can {LMs} Interpret Verbalized Uncertainty?}, author={Belem, Catarina G. and Kelly, Matthew and Singh, Sameer and Steyvers, Mark and Smyth, Padhraic}, booktitle={TrustNLP Workshop at NAACL}, year={2024}, note={Runner-Up Best Short Paper} } - International Conference on Learning Representations (ICLR 2024)PDF
@inproceedings{belem2024iclr, title={Are Models Biased on Text without Gender-related Language?}, author={Belem, Catarina G. and Seshadri, Shubha and Razeghi, Yasaman and Singh, Sameer}, booktitle={International Conference on Learning Representations}, year={2024} }
2023
- International Conference on Learning Representations (ICLR 2023)PDF
@inproceedings{cruz2023iclr, title={{FairGBM}: Gradient Boosting with Fairness Constraints}, author={Cruz, Andr{\'e} F. and Belem, Catarina G. and Bravo, Joao and Saleiro, Pedro and Bizarro, Pedro}, booktitle={International Conference on Learning Representations}, year={2023} }
2021
- IEEE International Conference on Data Mining (ICDM 2021)PDF
@inproceedings{cruz2021icdm, title={Promoting Fairness through Hyperparameter Optimization}, author={Cruz, Andr{\'e} F. and Saleiro, Pedro and Belem, Catarina G. and Soares, Carlos and Bizarro, Pedro}, booktitle={IEEE International Conference on Data Mining}, pages={1036--1041}, year={2021} } - Weakly Supervised Learning Workshop (WeaSul) at ICLR 2021PDF
@inproceedings{belem2021weasul, title={Weakly Supervised Multi-Task Learning for Concept-Based Explainability}, author={Belem, Catarina G. and Balayan, Varduhi and Saleiro, Pedro and Bizarro, Pedro}, booktitle={ICLR Workshop on Weakly Supervised Learning}, year={2021} } - How Can I Choose an Explainer? An Application-grounded Evaluation of Post-hoc Explanations ConferenceACM Conference on Fairness, Accountability, and Transparency (FAccT 2021)PDF
@inproceedings{jesus2021facct, title={How Can {I} Choose an Explainer? {An} Application-grounded Evaluation of Post-hoc Explanations}, author={Jesus, S{\'e}rgio and Belem, Catarina G. and Balayan, Varduhi and Bento, Jo{\~a}o and Saleiro, Pedro and Bizarro, Pedro and Gama, Jo{\~a}o}, booktitle={ACM Conference on Fairness, Accountability, and Transparency}, year={2021} }
2020
- Human And Machine in-the-Loop Evaluation and Learning Strategies (HAMLETS) Workshop at NeurIPS 2020PDF
@inproceedings{balayan2020hamlets, title={Teaching the Machine to Explain Itself using Domain Knowledge}, author={Balayan, Varduhi and Belem, Catarina G. and Saleiro, Pedro and Bizarro, Pedro}, booktitle={NeurIPS Workshop on Human And Machine in-the-Loop Evaluation and Learning Strategies}, year={2020} }