4 papers
Reassessing Global Gradient-Norm Imbalance in BLIP Fine-Tuning Across Physical Domains
Kiran Naseer, Samreen Azhar, Dwarikanath Mahapatra
Imbalanced gradient magnitudes between the visual and language pathways of a vision-language model are often treated as a defect to be corrected. We test that premise for one famil…
Perplexity Predicts Protection: Choosing Pretrained Backbones for Worst-Client Fairness in Federated Parameter-Efficient Fine-Tuning
Kiran Naseer, Samreen Azhar, Umar Shoaib +2
Federated learning lets multiple parties train a shared model without pooling their data, but a client with far less data than the others can end up poorly served even when the gro…
MedFL-Stress: A Systematic Robustness Evaluation of Federated Brain Tumor Segmentation under Cross-Hospital MRI Appearance Shift
Kiran Naseer, Naveed Anwer Butt
Federated learning lets hospitals train segmentation models without sharing patient data, but standard evaluation reports one number: mean Dice across all sites. That number can lo…
When More Parameters Hurt: Foundation Model Priors Amplify Worst-Client Disparity Under Extreme Federated Heterogeneity
Kiran Naseer, Umar Shoaib
Federated learning (FL) is increasingly used to fine-tune foundation models (FMs) on distributed private data. The community largely assumes that large-scale pretraining serves as…