6 papers
Least but not Last: Fine-tuning Intermediate Principal Components for Better Performance-Forgetting Trade-Offs
Alessio Quercia, Arya Bangun, Ira Assent +1
Low-Rank Adaptation (LoRA) methods have emerged as crucial techniques for adapting large pre-trained models to downstream tasks under computational and memory constraints. However,…
Classifier Reconstruction Through Counterfactual-Aware Wasserstein Prototypes
Xuan Zhao, Zhuo Cao, Arya Bangun +2
Counterfactual explanations provide actionable insights by identifying minimal input changes required to achieve a desired model prediction. Beyond their interpretability benefits,…
Physics-Guided Diffusion Priors for Multi-Slice Reconstruction in Scientific Imaging
Laurentius Valdy, Richard D. Paul, Alessio Quercia +4
Accurate multi-slice reconstruction from limited measurement data is crucial to speed up the acquisition process in medical and scientific imaging. However, it remains challenging…
FlowTIE: Flow-based Transport of Intensity Equation for Phase Gradient Estimation from 4D-STEM Data
Arya Bangun, Maximilian Töllner, Xuan Zhao +2
We introduce FlowTIE, a neural-network-based framework for phase reconstruction from 4D-Scanning Transmission Electron Microscopy (STEM) data, which integrates the Transport of Int…
1LoRA: Summation Compression for Very Low-Rank Adaptation
Alessio Quercia, Zhuo Cao, Arya Bangun +4
Parameter-Efficient Fine-Tuning (PEFT) methods have transformed the approach to fine-tuning large models for downstream tasks by enabling the adjustment of significantly fewer para…
MRI Reconstruction with Regularized 3D Diffusion Model (R3DM)
Arya Bangun, Zhuo Cao, Alessio Quercia +2
Magnetic Resonance Imaging (MRI) is a powerful imaging technique widely used for visualizing structures within the human body and in other fields such as plant sciences. However, t…