REVIEW 30 cited by
SVD-LLM: Truncation-aware Singular Value Decomposition for Large Language Model Compression
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
SVD-LLM: Truncation-aware Singular Value Decomposition for Large Language Model Compression
read the original abstract
The advancements in Large Language Models (LLMs) have been hindered by their substantial sizes, which necessitates LLM compression methods for practical deployment. Singular Value Decomposition (SVD) offers a promising solution for LLM compression. However, state-of-the-art SVD-based LLM compression methods have two key limitations: truncating smaller singular values may lead to higher compression loss, and the lack of update on the compressed weights after SVD truncation. In this work, we propose SVD-LLM, a SVD-based post-training LLM compression method that addresses the limitations of existing methods. SVD-LLM incorporates a truncation-aware data whitening technique to ensure a direct mapping between singular values and compression loss. Moreover, SVD-LLM adopts a parameter update with sequential low-rank approximation to compensate for the accuracy degradation after SVD compression. We evaluate SVD-LLM on 10 datasets and seven models from three different LLM families at three different scales. Our results demonstrate the superiority of SVD-LLM over state-of-the-arts, especially at high model compression ratios. Our code is available at https://github.com/AIoT-MLSys-Lab/SVD-LLM
Forward citations
Cited by 30 Pith papers
-
ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs
ENEC delivers 3.43X higher throughput than DietGPU and 1.12X better compression ratio than nvCOMP for lossless model weight compression on Ascend NPUs, yielding up to 6.3X end-to-end inference speedup.
-
Spectral Rewiring for Exploration, Purification, and Model Merging
Subspace-Aligned Rewiring projects RL weight updates onto the base model’s SVD basis, retaining a compact rewiring matrix that preserves reasoning and improves exploration and multi-domain merging.
-
LASER: Loss-Aware Singular-value Decomposition and Rank Allocation for Efficient Low-Precision Vision-Language Models
LASER introduces curvature-weighted SVD from second-order loss approximation and loss-aware rank allocation to compress VLMs, reporting over 2.3x decoding speedup under low-precision settings.
-
DREAM-S: Speculative Decoding with Searchable Drafting and Target-Aware Refinement for Multimodal Generation
DREAM-S combines neural architecture search, target-aware supernet training, and attention-entropy-guided distillation to accelerate speculative decoding in VLMs, reporting up to 3.85x speedup over standard methods.
-
SAFE-SVD: Sensitivity-Aware Fidelity-Enforcing SVD for Physics Foundation Models
SAFE-SVD introduces a sensitivity-aware fidelity-enforcing SVD framework for compressing physics foundation models that maintains higher accuracy than standard methods at greater compression ratios.
-
OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization
OSCAR achieves near-BF16 accuracy for 2-bit KV cache quantization by using offline spectral covariance-aware rotations aligned with attention, plus a custom deployable INT2 kernel compatible with paged serving.
-
Dynamic Model Merging Made Slim
DiDi-Merging achieves dynamic model merging performance matching or exceeding prior methods while using only 1.24x to 1.4x the parameters of a single fine-tuned model.
-
Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression
PARSE trains a prompt-aware linear router on dense-model outputs to select dynamic SVD ranks, improving accuracy up to 10% at 0.6 compression ratio on LLaMA-7B while delivering 2.5x prefill and 2.4x decode speedups.
-
Bayesian Fine-tuning in Projected Subspaces
Bayesian fine-tuning of large models can be done efficiently by projecting uncertainties into low-dimensional subspaces, yielding improved calibration and generalization while keeping computational costs low.
-
Gated Subspace Inference for Transformer Acceleration
Gated Subspace Inference accelerates transformer linear layers 3-10x via low-rank cached subspace computation and per-token gating to skip residuals while preserving output distribution to high accuracy.
-
Technical Report: Activation Residual Hessian Quantization (ARHQ) for Low-Bit LLM Quantization
ARHQ isolates error-sensitive weight directions in LLMs via truncated SVD on the scaled matrix W G_x^{1/2} from activation residuals, improving SNR and preserving performance under aggressive low-bit quantization.
-
SLaB: Sparse-Lowrank-Binary Decomposition for Efficient Large Language Models
SLaB compresses LLM weights via sparse-lowrank-binary decomposition guided by activation-aware scores, achieving up to 36% lower perplexity than prior methods at 50% compression on Llama models.
-
RUQuant: Towards Refining Uniform Quantization for Large Language Models
RUQuant uses block-wise composite orthogonal matrices from Householder reflections and Givens rotations plus a fine-tuned global reflection to achieve 99.8% full-precision accuracy at W6A6 and 97% at W4A4 for 13B LLMs...
-
Spectral Compact Training: Pre-Training Large Language Models via Permanent Truncated SVD and Stiefel QR Retraction
SCT pre-trains LLMs by keeping weights as compact SVD factors with Stiefel QR retraction, delivering up to 199x memory reduction per layer and allowing 70B-parameter training on a Steam Deck.
-
A3 : an Analytical Low-Rank Approximation Framework for Attention
A3 splits Transformer layers into QK, OV, and MLP components and derives analytical low-rank approximations that reduce hidden dimensions while minimizing each component's functional loss, yielding better perplexity t...
-
ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models
ASVD compresses LLMs by 10-30% and KV caches by 50% via activation-aware SVD that absorbs outliers into transformed weights and calibrates per-layer sensitivity.
-
Activation- and Influence-Aware Ranks (AIR): Function-Preserving SVD Compression for LLMs
AIR augments activation-aware SVD compression of LLMs with an influence metric and a closed-form ALS update, claiming >18% perplexity improvement at 60% parameter retention and 90% less calibration data than SVD-LLM(W).
-
EinSort: Sorting is All We Need for Tensorizing LLM
Sorting tensor indices enables an adaptive tensorization method that discovers low-rank structure in LLM weights and KV caches, yielding better reconstruction quality than baselines.
-
SigmaScale: LLM Compression with SVD-based Low-Rank Decomposition and Learned Scaling Matrices
Learned diagonal scaling matrices optimized with activation-aware loss reduce effective rank in LLM weight matrices and yield competitive perplexity and zero-shot results versus prior SVD methods on Llama 3.1 8B and Qwen3-8B.
-
Rethinking the Role of Tensor Decompositions in Post-Training LLM Compression
Tensor decompositions face practical limits in large-scale LLM compression due to mismatch between assumed shared subspaces and heterogeneous model representations.
-
Cross-Layer Subspace Coupling for LLM Compression: A Unifying Framework and Its Empirical Limits
Unifying cross-layer SVD compression for LLMs improves weight reconstruction error by up to 46% on Pythia models but causes severe degradation in perplexity and accuracy due to residual stream decoupling.
-
Fast Tensorization of Neural Networks via Slice-wise Feature Distillation
A slice-wise feature distillation framework for independent tensorization of neural network slices to achieve scalable compression with reduced fine-tuning costs.
-
Break Through the Compression Bottleneck: From Theory to Practice
The paper asserts a first proof that low-rank decomposition and quantization are non-orthogonal tools for LLM compression, recommends low-rank-first ordering, and adds a diagonal scaling fix (DAM) that reduces the com...
-
Activation-Informed Pareto-Guided Low-Rank Compression for Efficient LLM/VLM
A single per-layer error tolerance, instead of a per-layer rank budget, improves zero-shot low-rank compression of LLMs and VLMs empirically, but the theorem that this uniform tolerance is Pareto-optimal is proven onl...
-
PanoLora: Bridging Perspective and Panoramic Video Generation with LoRA Adaptation
Fine-tuning a pretrained video diffusion model with LoRA rank 16 on about 1,000 synthetic videos produces panoramic video with good seam closure, but the claim that rank must exceed 8 degrees of freedom is not proven.
-
ScatterPrism: convergence for generative simulation and inverse problems in particle and nuclear physics
CFM training loss plateaus while physics-informed metrics continue improving; ScatterPrism and a multi-metric protocol are proposed to restore kinematic fidelity without memorization.
-
ScatterPrism: convergence for generative simulation and inverse problems in particle and nuclear physics
Standard Conditional Flow Matching loss is a misleading early plateau; physics-informed metrics keep improving, so ScatterPrism and multi-metric diagnostics are needed for kinematic fidelity.
-
Accelerating Attention with Basis Decomposition
A low-rank factorization of attention projection matrices (basis plus coefficients) gives modest FLOP savings in exact arithmetic, but the claimed losslessness and novelty are not supported.
-
PHLoRA: data-free Post-hoc Low-Rank Adapter extraction from full-rank checkpoint
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.
-
A Survey on Efficient Inference for Large Language Models
The paper surveys techniques to speed up and reduce the resource needs of LLM inference, organized by data-level, model-level, and system-level changes, with comparative experiments on representative methods.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.