6 papers
DeconDTN-Toolkit: A Library for Evaluation and Enhancement of Robustness to Provenance Shift
Yongsen Tan, Zhecheng Sheng, Xiruo Ding +2
Despite the burgeoning body of work on distribution shifts, provenance shift-where the relationship between data source and label changes at deployment-remains poorly understood an…
When Prompts Interact: Assessing Prompt Arithmetic for Deconfounding under Distribution Shift
Zhecheng Sheng, Yongsen Tan, Xiruo Ding +2
In classification tasks, models may rely on confounding variables to achieve strong in-distribution performance, capturing spurious features that fail under distribution shift. Thi…
Mitigating Confounding in Speech-Based Dementia Detection through Weight Masking
Zhecheng Sheng, Xiruo Ding, Brian Hur +3
Deep transformer models have been used to detect linguistic anomalies in patient transcripts for early Alzheimer's disease (AD) screening. While pre-trained neural language models…
"Is There Anything Else?'': Examining Administrator Influence on Linguistic Features from the Cookie Theft Picture Description Cognitive Test
Changye Li, Zhecheng Sheng, Trevor Cohen +1
Alzheimer's Disease (AD) dementia is a progressive neurodegenerative disease that negatively impacts patients' cognitive ability. Previous studies have demonstrated that changes in…
Reexamining Racial Disparities in Automatic Speech Recognition Performance: The Role of Confounding by Provenance
Changye Li, Trevor Cohen, Serguei Pakhomov
Automatic speech recognition (ASR) models trained on large amounts of audio data are now widely used to convert speech to written text in a variety of applications from video capti…
Too Big to Fail: Larger Language Models are Disproportionately Resilient to Induction of Dementia-Related Linguistic Anomalies
Changye Li, Zhecheng Sheng, Trevor Cohen +1
As artificial neural networks grow in complexity, understanding their inner workings becomes increasingly challenging, which is particularly important in healthcare applications. T…