PATCH: Learnable Tile-level Hybrid Sparsity for LLMs

arXiv:2509.23410 · cs.LG, cs.AI, cs.PF · Submitted 2025-09-27 · Read on arXiv

cs.LG, cs.AI, cs.PF

Submitted: 2025-09-27

Updated: 2026-10-05

Code: https://github.com/Paramathic/patch

Project page: https://paramathic.github.io/stoicc-docs

Terminology

• Sparsity — 6× in this paper · explained in 4 episodes
• Accuracy — 4× in this paper · explained in 1 episode
• Pruning — 2× in this paper · explained in 4 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes

Sources

• Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• MaskLLM: Learnable Semi-Structured Sparsity for Large Language Models
• The Llama 3 Herd of Models
• LQ-LoRA: Low-rank Plus Quantized Matrix Decomposition for Efficient Language Model Finetuning
• Thanos: A Block-wise Pruning Algorithm for Efficient Large Language Model Compression
• Categorical Reparameterization with Gumbel-Softmax
• Layer-adaptive sparsity for the Magnitude-based Pruning
• ProxSparse: Regularized Learning of Semi-Structured Sparsity Masks for Pretrained LLMs
• Decoupled Weight Decay Regularization
• AffineQuant: Affine Transformation Quantization for Large Language Models
• Pointer Sentinel Mixture Models
• Accelerating Sparse Deep Neural Networks
• Compression Trinity: Exploring Sparsity, Quantization, and Low-Rank Approximations for LLM Compression
• OPTIMA: Optimal One-shot Pruning for LLMs via Quadratic Programming Reconstruction
• LEAP: Learnable End-to-End Adaptive Pruning of Large Language Models
• WinoGrande: An Adversarial Winograd Schema Challenge at Scale
• A Simple and Effective Pruning Approach for Large Language Models
• Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them

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