Publications

1. The Multilingual Alignment Paradox: Preference Conflict Hurts Alignment but Prevents Language Collapse.
   Binwei Yao, Zhengguang Zhou, Rheeya Uppaal, Junjie Hu
   Preprint, 2026.
2. Whose Values Are Generated? Measuring Visual Value Leakage in Multilingual Text-to-Image Generation.
   Zucan Lyu, Binwei Yao, Rheeya Uppaal, Jing Yao, Xiaoyuan Yi, Junjie Hu
   Preprint, 2026.
3. OpenSafeIntent: Evaluating Intent-Calibrated Safe Completion Across Dual-Use Prompt Sets.
   Rheeya Uppaal, Seungwoo Lyu, Selina Sung, Junjie Hu
   Preprint, 2026.
4. When Safety Fails Before the Answer: Benchmarking Harmful Behavior Detection in Reasoning Chains.
   Ishita Kakkar, Enze Zhang, Rheeya Uppaal, Junjie Hu
   In the Conference On Language Modeling (COLM), 2026.
5. Shattered Compositionality: Counterintuitive Learning Dynamics of Transformers for Arithmetic.
   Xingyu Zhao, Darsh Sharma, Rheeya Uppaal, Yiqiao Zhong
   In the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS), 2026.
6. Journey Before Destination: On the importance of Visual Faithfulness in Slow Thinking. [Oral]
   Rheeya Uppaal, Phu Mon Htut, Min Bai, Nikolaos Pappas, Zheng Qi, Sandesh Swamy
   In the 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL), 2026 (Oral Presentation).
7. Model Editing as a Robust and Denoised variant of DPO: A Case Study on Toxicity. [Talk]
   Rheeya Uppaal, Apratim Dey, Yiting He, Yiqiao Zhong, Junjie Hu
   In the Thirteenth International Conference on Learning Representations (ICLR), 2025.
   *A non-archival version was presented at the Safe Generative AI Workshop at the Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS), 2024.
8. Improving Bilingual Capabilities of Language Models to Support Diverse Linguistic Practices in Education.
   Anand Syamkumar, Nora Tseng, Kaycie Barron, Shanglin Yang, Shamya Karumbaiah, Rheeya Uppaal, Junjie Hu.
   Preprint, 2024.
9. How Useful is Continued Pre-Training for Generative Unsupervised Domain Adaptation?
   Rheeya Uppaal, Yixuan Li, Junjie Hu
   In the 9th Workshop on Representation Learning for NLP (Rep4NLP) at the 62nd Annual Meeting of the Association for Computational Linguistics (ACL), 2024.
10. Evolving Domain Adaptation of Pretrained Language Models for Text Classification.
    Yun-Shiuan Chuang, Yi Wu, Dhruv Gupta, Rheeya Uppaal, Ananya Kumar, Luhang Sun, Makesh Narsimhan Sreedhar, Sijia Yang, Timothy T. Rogers, Junjie Hu
    In the Workshop on Distribution Shifts at the Thirty-seventh Conference on Neural Information Processing Systems (NeurIPS), 2023.
11. Is Fine-tuning Needed? Pre-trained Language Models Are Near Perfect for Out-of-Domain Detection. [Covered in media by CASMI] [Talk] [Code]
    Rheeya Uppaal, Junjie Hu, Yixuan Li.
    In the 61st Annual Meeting of the Association for Computational Linguistics (ACL), 2023.
12. Long Document Summarization in a Low Resource Setting using Pretrained Language Models.
    Ahsaas Bajaj, Pavitra Dangati, Kalpesh Krishna, Pradhiksha Ashok Kumar, Rheeya Uppaal, Bradford Windsor, Eliot Brenner, Dominic Dotterrer, Rajarshi Das, Andrew McCallum
    In the Student Workshop at the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP), 2021.
13. Overcoming Practical Issues of Deep Active Learning and its Applications on Named Entity Recognition.
    Haw-Shiuan Chang, Shankar Vembu, Sunil Mohan, Rheeya Uppaal, Andrew McCallum.
    In the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2020 (Journal Track)
14. LRS-DAG: Low Resource Supervised Domain Adaptation with Generalization Across Domains. [Talk]
    Rheeya Uppaal
    In the New in ML Workshop of the Thirty-third Conference on Neural Information Processing Systems (NeurIPS), 2019.
15. Multi-resolution Attention with Signal Splitting for Multivariate Time Series Classification.
    Rheeya Uppaal, Bryon Kucharski, Bhanu Pratap Singh, Iman Deznabi, Madelina Fiterau.
    Time Series Workshop of The Thirty-sixth International Conference on Machine Learning (ICML), 2019.
16. Multi-resolution Networks For Flexible Irregular Time Series Modeling (Multi-FIT)
    Bhanu Pratap Singh, Iman Deznabi, Rheeya Uppaal, Bryon Kucharski, Bharath Narasimhan, Akhila Josyula, Madelina Fiterau.
    Preprint.
17. Leveraging One's Impact Through Digital Persona and Image Management for Enhanced Effectiveness and Competitive Advantage.
    Kuiljeit Uppaal, Rheeya Uppaal
    Elsevier, 2018.
18. Exploring Robust Vehicle Detection and Tracking Methods for Various Illumination Settings using Infrared Thermographic Images
    Rheeya Uppaal.
    In the International Journal of Research in Advent Technology, Special Issue National Conference NCPC, 2016.
19. Exploring the Facets of Colonization of the Red Planet – Mars.
    Rheeya Uppaal
    In the Student Conference on Emerging Technologies and Innovations, 2014.