NewA calmer, faster ZeroHour — security assessments on articles, story timelines and search from any page.

#post-training-quantization

0 stories this week · 1 this month · 2 in 6 months · mostly AI

Stories per day · 30 days

Relateddiffusion1 drunet1 mixed-precision1 model-compression1 quantization1

2 stories, newest first · one card per story

arXiv cs.AI / cs.LG / cs.CL24d ago

Layerwise Error Attribution for Fast and Robust Mixed-Precision Post-Training Quantization

Layerwise error scores speed mixed-precision quantization up to 2,570x and resist corrupted calibration data.

quantization mixed-precision

1

Hugging Face daily papersSep 28, 2026

STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization

STEPQuant quantizes delta-rule recurrent states by error lifetime and location, matching FP32 accuracy at 6 bits.

quantization linear-attention

2