Squeeze10-LLM: Squeezing LLMs' Weights by 10 Times via a Staged Mixed-Precision Quantization Method

arXiv:2507.18073 · cs.LG · Submitted 2025-07-24 · Read on arXiv

cs.LG

Submitted: 2025-07-24

Updated: 2026-09-08

Code: https://github.com/EleutherAI/lmevaluation-harness

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Quantization — 132× in this paper · explained in 12 episodes
• Accuracy — 22× in this paper · explained in 1 episode
• Perplexity — 9× in this paper · explained in 3 episodes
• Robustness — 9× in this paper · explained in 5 episodes
• Adam — 7× in this paper · explained in 2 episodes
• Full Information Activation Supervision — 7× in this paper
• Core — 6× in this paper · explained in 2 episodes
• Post-Binarization Activation Robustness — 5× in this paper
• Transformer — 4× in this paper · explained in 2 episodes
• Elias — 3× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 3× in this paper · explained in 6 episodes
• Loss Function — 3× in this paper · explained in 2 episodes
• Benchmark — 2× in this paper · explained in 2 episodes
• Institutional Review Board — 2× in this paper
• Sparsity — 2× in this paper · explained in 4 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Ablation study — 1× in this paper · explained in 1 episode
• Clustering — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes
• Pruning — 1× in this paper · explained in 4 episodes
• Trace — 1× in this paper · explained in 2 episodes
• Transfer Learning — 1× in this paper · explained in 10 episodes

Sources

• GPT-4 Technical Report
• Progressive Mixed-Precision Decoding for Efficient LLM Inference
• BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models
• GPT-4o System Card
• OpenAI o1 System Card
• DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
• DeepSeek-V3 Technical Report
• DeepSeek-VL: Towards Real-World Vision-Language Understanding
• Pointer Sentinel Mixture Models
• PB-LLM: Partially Binarized Large Language Models
• OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models
• LLaMA: Open and Efficient Foundation Language Models
• Llama 2: Open Foundation and Fine-Tuned Chat Models
• RPTQ: Reorder-based Post-training Quantization for Large Language Models
• HellaSwag: Can a Machine Really Finish Your Sentence?
• MixLLM: LLM Quantization with Global Mixed-precision between Output-features and Highly-efficient System Design

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