Pith. sign in

Title resolution pending

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.CL 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

L1RA: Dynamic Rank Assignment in LoRA Fine-Tuning

cs.CL · 2025-09-05 · conditional · novelty 5.0

L1RA uses L1 regularization to prune and reallocate LoRA ranks across adapters during fine-tuning, achieving small perplexity gains on two 7B to 8B models at nearly the same training time.

citing papers explorer

Showing 1 of 1 citing paper.

  • L1RA: Dynamic Rank Assignment in LoRA Fine-Tuning cs.CL · 2025-09-05 · conditional · none · ref 24

    L1RA uses L1 regularization to prune and reallocate LoRA ranks across adapters during fine-tuning, achieving small perplexity gains on two 7B to 8B models at nearly the same training time.