5 papers · 1 filter
Measuring Pretraining Contamination in Public Medical Vision-Language Benchmarks
Bruce Changlong Xu, Lan Wu, Alexander Ryu
Medical vision-language models (VLMs) are evaluated on public benchmarks whose images and question-answer pairs have been freely available for years, yet reported accuracy assumes…
Alignment Collapse Under KV Cache Quantization: Diagnosis and Mitigation
Bruce Changlong Xu, Adarsh Kumarappan, Mu Zhou
Key-value (KV) cache quantization is widely used to reduce Large Language Model (LLM) inference memory, yet existing evaluations solely focus on measuring perplexity and accuracy w…
Behavioral Consistency in Machine Learning
Bruce Changlong Xu, Lan Wu
A trained model is usually described by its weights, as if the weights fixed its behavior. In practice the same weights are run by different programs. Training uses high-precision…
From Theory to Decision Rule: Calibrating the Noisy-Label Crossover for Vision-Language Model Weak Supervision Across Three Medical-Imaging Benchmarks
Bruce Changlong Xu, Jose James, Jose K. James +2
In medical imaging, small sets of expert labels are often combined with zero-shot pseudo-labels from a vision-language model (VLM). When the expert-labeled set is very small, the e…
Activation Sensitivity as a Unifying Principle for Post-Training Quantization
Bruce Changlong Xu
Post-training quantization (PTQ) methods for large language models compress pretrained weights to low bit-widths without retraining by optimizing layer-local proxy objectives. Acti…