# Surgical Feature-Space Decomposition of LLMs: Why, When and How?

arXiv: https://arxiv.org/abs/2405.13039 · HTML: https://mlsanity.com/paper/2405.13039/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 45 references that are in the map.
- First posted on arXiv 17 May 2024.

## Builds on (most cited first, 30 of 31)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://mlsanity.com/paper/1803.03635/) (2018)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems](https://mlsanity.com/paper/1905.00537/) (2019)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes](https://mlsanity.com/paper/2305.02301/) (2023)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Rethinking the Value of Network Pruning](https://mlsanity.com/paper/1810.05270/) (2018)
- [Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation](https://mlsanity.com/paper/1404.0736/) (2014)
- [Speeding up Convolutional Neural Networks with Low Rank Expansions](https://mlsanity.com/paper/1405.3866/) (2014)
- [Surgical Fine-Tuning Improves Adaptation to Distribution Shifts](https://mlsanity.com/paper/2210.11466/) (2022)
- [A Tutorial on Principal Component Analysis](https://mlsanity.com/paper/1404.1100/) (2014)
- [Accelerating Very Deep Convolutional Networks for Classification and Detection](https://mlsanity.com/paper/1505.06798/) (2015)
- [LoSparse: Structured Compression of Large Language Models based on Low-Rank and Sparse Approximation](https://mlsanity.com/paper/2306.11222/) (2023)
- [Stable Low-rank Tensor Decomposition for Compression of Convolutional Neural Network](https://mlsanity.com/paper/2008.05441/) (2020)
- [Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning](https://mlsanity.com/paper/2102.03983/) (2021)
- [Vision Transformer Slimming: Multi-Dimension Searching in Continuous Optimization Space](https://mlsanity.com/paper/2201.00814/) (2022)
- [Gold Doesn't Always Glitter: Spectral Removal of Linear and Nonlinear Guarded Attribute Information](https://mlsanity.com/paper/2203.07893/) (2022)
- [Efficient fair PCA for fair representation learning](https://mlsanity.com/paper/2302.13319/) (2023)
- [ChipNet: Budget-Aware Pruning with Heaviside Continuous Approximations](https://mlsanity.com/paper/2102.07156/) (2021)
- [Understanding the Effect of Model Compression on Social Bias in Large Language Models](https://mlsanity.com/paper/2312.05662/) (2023)
- [Layer-wise Auto-Weighting for Non-Stationary Test-Time Adaptation](https://mlsanity.com/paper/2311.05858/) (2023)
- [On Surgical Fine-tuning for Language Encoders](https://mlsanity.com/paper/2310.17041/) (2023)
- [Erasure of Unaligned Attributes from Neural Representations](https://mlsanity.com/paper/2302.02997/) (2023)

## Built on by (most cited first, 2 of 2)

- [GRASP: Replace Redundant Layers with Adaptive Singular Parameters for Efficient Model Compression](https://mlsanity.com/paper/2501.00339/) (2024)
- [CAPA: Contribution-Aware Pruning and FFN Approximation for Efficient Large Vision-Language Models](https://mlsanity.com/paper/2602.00247/) (2026)

## Abstract (arXiv)

Low-rank approximations, of the weight and feature space can enhance the performance of deep learning models, whether in terms of improving generalization or reducing the latency of inference. However, there is no clear consensus yet on how, when and why these approximations are helpful for large language models (LLMs). In this work, we empirically study the efficacy of weight and feature space decomposition in transformer-based LLMs. We demonstrate that surgical decomposition not only provides critical insights into the trade-off between compression and language modelling performance, but also sometimes enhances commonsense reasoning performance of LLMs. Our empirical analysis identifies specific network segments that intrinsically exhibit a low-rank structure. Furthermore, we extend our investigation to the implications of low-rank approximations on model bias. Overall, our findings offer a novel perspective on optimizing LLMs, presenting the low-rank approximation not only as a tool for performance enhancements, but also as a means to potentially rectify biases within these models. Our code is available at hrefhttps://github.com/nyunAI/SFSD-LLMGitHub.