8 papers
Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification
Kimia Hamidieh, Veronika Thost, Walter Gerych +2
Large language models (LLMs) often produce confident yet incorrect responses, and uncertainty quantification is one potential solution to more robust usage. Recent works routinely…
When Style Breaks Safety: Defending LLMs Against Superficial Style Alignment
Yuxin Xiao, Sana Tonekaboni, Walter Gerych +2
Large language models (LLMs) can be prompted with specific styles (e.g., formatting responses as lists), including in malicious queries. Prior jailbreak research mainly augments th…
An Investigation of Memorization Risk in Healthcare Foundation Models
Sana Tonekaboni, Lena Stempfle, Adibvafa Fallahpour +2
Foundation models trained on large-scale de-identified electronic health records (EHRs) hold promise for clinical applications. However, their capacity to memorize patient informat…
The MedPerturb Dataset: What Non-Content Perturbations Reveal About Human and Clinical LLM Decision Making
Abinitha Gourabathina, Yuexing Hao, Walter Gerych +1
Clinical robustness is critical to the safe deployment of medical Large Language Models (LLMs), but key questions remain about how LLMs and humans may differ in response to the rea…
Learning under Temporal Label Noise
Sujay Nagaraj, Walter Gerych, Sana Tonekaboni +3
Many time series classification tasks, where labels vary over time, are affected by label noise that also varies over time. Such noise can cause label quality to improve, worsen, o…
MaskMedPaint: Masked Medical Image Inpainting with Diffusion Models for Mitigation of Spurious Correlations
Qixuan Jin, Walter Gerych, Marzyeh Ghassemi
Spurious features associated with class labels can lead image classifiers to rely on shortcuts that don't generalize well to new domains. This is especially problematic in medical…