SOURCE-LINKED INTELLIGENCE
Understanding LLM Quantization through Activation-Guided Compensation and Orthogonal Residuals
Post-training weight-activation quantization reduces the memory and inference costs of large language models, but aggressive W4A4 quantization remains difficult because activation outliers degrade effective quantization resolution. Although weight optimization, channel-wise scaling, and orthogonal rotation mitigate this problem, the error components they address and their relationship remain unclear. Using an exact decomposition of local weight-activation quantization error into an activation-guided weight compensation term and an orthogonal residual, we bound the residual using persistent cha
Read original source ↗ Open in workspace
- recordType
- paper
- region
- Global
Evidence & attribution
- arXiv · AI, language, vision and robotics · 2026-09-18T08:00:27.000Z
First collected: 2026-09-23T14:01:59.594Z. This is not the publication date.