Tensorization is a powerful but underexplored tool for compression and interpretability of neural networks
cs.LG, cs.AI, quant-ph
Submitted: 2025-05-26
Updated: 2026-09-14
Comments: This article has been prepared for submission as a "Position paper" following the guidelines provided at https://neurips.cc/Conferences/2025/CallForPositionPapers
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Scaling Laws for Neural Language Models
- Can Neural Network Memorization Be Localized?
- The Super Weight in Large Language Models
- DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
- LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
- A Simple and Effective Pruning Approach for Large Language Models
- Distilling the Knowledge in a Neural Network
- Model compression via distillation and quantization
- Tensor Networks Meet Neural Networks: A Survey and Future Perspectives
- Era of Big Data Processing: A New Approach via Tensor Networks and Tensor Decompositions
- Boosting Defect Detection in Manufacturing using Tensor Convolutional Neural Networks
- Tensor network compressibility of convolutional models
- CompactifAI: Extreme Compression of Large Language Models using Quantum-Inspired Tensor Networks
- TensorGPT: Efficient Compression of Large Language Models based on Tensor-Train Decomposition
- ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models
- Quantum Large Language Models via Tensor Network Disentanglers
- Scaling Laws vs Model Architectures: How does Inductive Bias Influence Scaling?
- LLaMA: Open and Efficient Foundation Language Models
- Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition
- Sparse Autoencoders Find Highly Interpretable Features in 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: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks