PACE quantizes linear-attention recurrent states to four bits post hoc
Mixed four- and eight-bit PACE matches eight-bit DAMP on the three larger hybrid models at about two thirds of its memory.
4 stories, newest first · one card per story
Mixed four- and eight-bit PACE matches eight-bit DAMP on the three larger hybrid models at about two thirds of its memory.
Theory shows a fixed recurrent memory matching current predictions can still be insufficient for future learning.
Golden-ratio Weyl dither keeps quantized Mamba recurrent states closer to full precision than stochastic rounding.
STEPQuant quantizes delta-rule recurrent states by error lifetime and location, matching FP32 accuracy at 6 bits.
Today 1.7K readers · 3.7K views · 2.8K reads · all time 161K views