AI Paper Atlas

Research map / Architectures and efficiency / Quantization

SAES-SVD: Self-Adaptive Suppression of Accumulated and Local Errors for SVD-based LLM Compression

Xing Hu, Dawei Yang, Yuan Cheng, Zhixuan Chen, Zukang Xu

ICLR 2026 · Architectures and efficiency › Quantization · approach: KV cache quantization · 8 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add cumulative error-aware compensation (CEALC) (ablation)whitened SVD compression · WikiText2 PPL, LLaMA2-7Bperplexity: 9.34 -> 7.66Improved
add cumulative error-aware compensation (CEALC) (ablation)whitened SVD compression · WikiText2 PPL, LLaMA-3-8Bperplexity: 16.59 -> 12.25Improved
add adaptive coefficient selection (ACES) on top of CEALC (ablation)SAES-SVD · WikiText2 PPL, LLaMA2-7Bperplexity: 7.66 -> 7.37Improved
add ACES on top of CEALC (ablation)SAES-SVD · Avg accuracy, LLaMA-3-8Baccuracy (%): 58.82 -> 60.18Improved
restrict adaptive alpha range to [0.25, 0.75] rather than [0,1] (ablation)ACES · WikiText2 PPL, LLaMA2-7B ratio 0.2perplexity: 8.96 -> 7.37Improved
narrow alpha range to [0.4, 0.6] (ablation)ACES · WikiText2 PPL, LLaMA2-7B ratio 0.2perplexity: 7.37 -> 7.63Hurt
use SAES-SVD instead of SVD-LLM at high compressionSAES-SVD · WikiText2 PPL, LLaMA-3-8B ratio 0.8perplexity: 6971.65 -> 181.84Improved
use SAES-SVD instead of Dip-SVD with no fine-tuning or mixed rankSAES-SVD · Avg accuracy drop, LLaMA-7B ratio 0.2accuracy drop (%): 9.2% -> 3.9%Improved
add ASVD mixed-rank allocation to SAES-SVD (ablation)SAES-SVD · WikiText2 PPL, LLaMA-7B ratio 0.4perplexity: 10.42 -> 7.48Improved
SAES-SVD vs Dip-SVD on 7-task accuracy at ratio 0.4 (7B)SAES-SVD · Avg accuracy drop, LLaMA-7B ratio 0.4accuracy drop (%): 22.8% -> 21.1%Improved
SAES-SVD versus structured pruning at 7GBSAES-SVD · WikiText2 PPL, LLaMA-7B 7GBperplexity: 43.05 -> 10.15Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy