12 papers
Hair-Trigger Alignment: Black-Box Evaluation Cannot Guarantee Post-Update Alignment
Yavuz Bakman, Duygu Nur Yaldiz, Eleni Triantafillou +3
Large Language Models (LLMs) are rarely static and are frequently updated in practice. A growing body of alignment research has shown that models initially deemed ``aligned'' can e…
Balancing Classification and Calibration Performance in Decision-Making LLMs via Calibration Aware Reinforcement Learning
Duygu Nur Yaldiz, Evangelia Spiliopoulou, Zheng Qi +3
Large language models (LLMs) are increasingly deployed in decision-making tasks, where not only accuracy but also reliable confidence estimates are essential. Well-calibrated confi…
Conformal Prediction Adaptive to Unknown Subpopulation Shifts
Nien-Shao Wang, Duygu Nur Yaldiz, Yavuz Faruk Bakman +1
Conformal prediction is widely used to equip black-box machine learning models with uncertainty quantification, offering formal coverage guarantees under exchangeable data. However…
Reject Only Critical Tokens: Pivot-Aware Speculative Decoding
Amir Ziashahabi, Yavuz Faruk Bakman, Duygu Nur Yaldiz +3
Speculative Decoding (SD) ensures that the output matches the target model's distribution exactly. However, we argue that this distribution matching requirement is too stringent an…
Uncertainty as Feature Gaps: Epistemic Uncertainty Quantification of LLMs in Contextual Question-Answering
Yavuz Bakman, Sungmin Kang, Zhiqi Huang +8
Uncertainty Quantification (UQ) research has primarily focused on closed-book factual question answering (QA), while contextual QA remains unexplored, despite its importance in rea…
Uncertainty Quantification for Hallucination Detection in Large Language Models: Foundations, Methodology, and Future Directions
Sungmin Kang, Yavuz Faruk Bakman, Duygu Nur Yaldiz +2
The rapid advancement of large language models (LLMs) has transformed the landscape of natural language processing, enabling breakthroughs across a wide range of areas including qu…