IBDR couples Bayesian LoRA fine-tuning with a diversity-promoting divergence loss and Wasserstein distributional robustness, improving average ensemble accuracy on VTAB-1K and commonsense reasoning benchmarks.
Robust wasserstein profile inference and applications to machine learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Promoting Ensemble Diversity with Interactive Bayesian Distributional Robustness for Fine-tuning Foundation Models
IBDR couples Bayesian LoRA fine-tuning with a diversity-promoting divergence loss and Wasserstein distributional robustness, improving average ensemble accuracy on VTAB-1K and commonsense reasoning benchmarks.