[COLM 2025] DFRot: Achieving Outlier-Free and Massive Activation-Free for Rotated LLMs with Refined Rotation; 知乎:https://zhuanlan.zhihu.com/p/12186430182
-
Updated
Mar 5, 2025 - Python
[COLM 2025] DFRot: Achieving Outlier-Free and Massive Activation-Free for Rotated LLMs with Refined Rotation; 知乎:https://zhuanlan.zhihu.com/p/12186430182
Weight quantization
High-performance LLM compression engine using SVD matrix decomposition, INT8 quantization, and output caching training acceleration.
LLM Quantization Gallery source data
Practical experiments applying modern geometry, topology, algebra, probability and optimization to improve RAG, retrieval, fine-tuning and LLM inference.
H-matrices with integer arithmetic
To associate your repository with the quantization-algorithm topic, visit your repo's landing page and select "manage topics."