Pith. sign in

REVIEW 1 cited by

FlexiGPT: Pruning and Extending Large Language Models with Low-Rank Weight Sharing

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 2501.14713 v2 pith:QEAEAKVN submitted 2025-01-24 cs.CL cs.LG

classification cs.CLcs.LG
keywords performancebenchmarkslanguagelow-rankmodelsblockscompressiondemonstrate
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The rapid proliferation of large language models (LLMs) in natural language processing (NLP) has created a critical need for techniques that enable efficient deployment on memory-constrained devices without compromising performance. We present a method to prune LLMs that selectively prunes model blocks based on an importance score and replaces them with a low-parameter replacement strategy. Specifically, we propose a principled metric to replace each pruned block using a weight-sharing mechanism that leverages unpruned counterparts from the model and block-specific low-rank adapters. Furthermore, we facilitate the learning of these replacement blocks with output feature normalization and an adapter initialization scheme built on low-rank SVD reconstructions. Empirical evaluations demonstrate substantial performance gains over existing methods, achieving state-of-the-art performance on 5/6 benchmarks for a compression rate of 30% and 6/6 benchmarks for a compression rate of 40%. We also demonstrate that our approach can extend smaller models, boosting performance on 6/6 benchmarks using only ~0.3% tokens of extended training with minimal additional parameter costs.

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. CALR: Corrective Adaptive Low-Rank Decomposition for Efficient Large Language Model Layer Compression

    cs.LG 2025-08 unverdicted novelty 4.0 of 10

    CALR adds a parallel learnable low-rank module to recover functional performance lost by SVD-based compression of LLM weight matrices.

Pith tools