REVIEW 3 cited by
LoRETTA: Low-Rank Economic Tensor-Train Adaptation for Ultra-Low-Parameter Fine-Tuning 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
abstract
Various parameter-efficient fine-tuning (PEFT) techniques have been proposed to enable computationally efficient fine-tuning while maintaining model performance. However, existing PEFT methods are still limited by the growing number of trainable parameters with the rapid deployment of Large Language Models (LLMs). To address this challenge, we present LoRETTA, an ultra-parameter-efficient framework that significantly reduces trainable parameters through tensor-train decomposition. Specifically, we propose two methods, named {LoRETTA}$_{adp}$ and {LoRETTA}$_{rep}$. The former employs tensorized adapters, offering a high-performance yet lightweight approach for the fine-tuning of LLMs. The latter emphasizes fine-tuning via weight parameterization with a set of small tensor factors. LoRETTA achieves comparable or better performance than most widely used PEFT methods with up to $100\times$ fewer parameters on the LLaMA-2-7B models. Furthermore, empirical results demonstrate that the proposed method effectively improves training efficiency, enjoys better multi-task learning performance, and enhances the anti-overfitting capability. Plug-and-play codes built upon the Huggingface framework and PEFT library will be released.
Forward citations
Cited by 3 Pith papers
-
MetaTT: A Global Tensor-Train Adapter for Parameter-Efficient Fine-Tuning
MetaTT factorizes all transformer adapter weights into a single shared tensor-train, achieving LoRA-competitive accuracy with up to 30-40x fewer trainable parameters and a DMRG-inspired rank-adaptive optimizer.
-
ACE: Exploring Activation Cosine Similarity and Variance for Accurate and Calibration-Efficient LLM Pruning
ACE adds activation cosine-similarity and activation-variance terms to the per-weight importance score, and reports better perplexity and lower pruning time than Wanda and RIA on LLaMA, LLaMA-2, and OPT.
-
Mixture of Low Rank Adaptation with Partial Parameter Sharing for Time Series Forecasting
MoLA adapts a pre-trained short-horizon forecaster to multiple forecast steps via segment-specific mixtures of shared low-rank adapters, reporting modest mean-squared-error gains over the base models on most of eight ...
Discussion (0). Continue with ORCID to comment.