Zero Sum SVD: Balancing Loss Sensitivity for Low Rank LLM Compression

arXiv:2602.02848 · cs.LG · Submitted 2026-02-02 · Read on arXiv

cs.LG

Submitted: 2026-02-02

Updated: 2026-09-28

Code: https://github.com/mint-vu/Zero-Sum-SVD

Terminology

• Benchmark — 1× in this paper · explained in 2 episodes
• Calibration — 1× in this paper · explained in 7 episodes

Sources

• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• Everybody Prune Now: Structured Pruning of LLMs with only Forward Passes
• DipSVD: Dual-importance Protected SVD for Efficient LLM Compression
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• OpenAI GPT-5 System Card
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• A Simple and Effective Pruning Approach for Large Language Models
• ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks