Pith. sign in

REVIEW 1 cited by

SORSA: Singular Values and Orthonormal Regularized Singular Vectors Adaptation of Large Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2409.00055 v6 pith:BITIMD2N submitted 2024-08-21 cs.LG cs.CL

SORSA: Singular Values and Orthonormal Regularized Singular Vectors Adaptation of Large Language Models

classification cs.LG cs.CL
keywords sorsasingularanalyzeorthonormalweightsadaptationdiagefficient
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

In this paper, we propose Singular Values and Orthonormal Regularized Singular Vectors Adaptation, or SORSA, a novel parameter efficient fine-tuning (PEFT) method. Each SORSA adapter consists of two main parts: trainable principal singular weights $W_p = U_p \text{diag}(S_p) V^\top_p$, and frozen residual weights $W_r = U_r \text{diag}(S_r) V^\top_r$. These parts are initialized by performing singular value decomposition (SVD) on pre-trained weights. Moreover, we implement and analyze an orthonormal regularizer, which we prove could decrease the condition number of $W_p$ and make the optimization more efficient. SORSA adapters could be merged during inference, thus eliminating any inference latency. We also introduce a method to analyze the variation of the parameters by performing SVD and discuss and analyze SORSA's superiority in minimizing the alteration in the SVD aspect. After all, SORSA shows a faster convergence than LoRA and PiSSA in our experiments. On the GSM-8K benchmark, Llama 2 7B adapted using SORSA achieved 56.03\% accuracy, surpassing LoRA (42.30\%) and Full FT (49.05\%). We conclude that SORSA offers a new perspective on parameter-efficient fine-tuning, demonstrating remarkable performance.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. PHLoRA: data-free Post-hoc Low-Rank Adapter extraction from full-rank checkpoint

    cs.LG 2025-09 conditional novelty 3.0

    PHLoRA extracts LoRA-compatible adapters from full-rank fine-tuned models via truncated SVD of the weight delta, matching full-rank performance on several benchmarks with no gradients or training data.