BaRA adds Bayesian adaptive rank allocation to LoRA fine-tuning by activating sparse instance-specific latent factors, with a generalization bound depending on learned joint effective rank rather than fixed maximum rank.
Knowledge entropy decay during language model pretraining hinders new knowledge acquisition,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
BaRA: Bayesian Adaptive Rank Allocation for Parameter-Efficient Fine-Tuning
BaRA adds Bayesian adaptive rank allocation to LoRA fine-tuning by activating sparse instance-specific latent factors, with a generalization bound depending on learned joint effective rank rather than fixed maximum rank.