For-Value: Efficient Forward-Only Data Valuation for finetuning LLMs and VLMs

For-Value estimates training-data influence from a single forward pass, avoiding gradient computations for scalable LLM and VLM data valuation.

A Closer Look at Personalized Fine-Tuning in Heterogeneous Federated Learning

A study of federated feature distortion shows how linear probing followed by fine-tuning can improve the balance between client personalization and generalization.

DARE the Extreme: Revisiting Delta-Parameter Pruning For Fine-Tuned Models

DARq and AdamR improve delta-parameter pruning under high pruning rates or large parameter changes for efficient fine-tuned model storage and deployment.