dSpark distills unmasking schedules to speed diffusion language models
On GSM8K the paper reports 10.12× and 5.12× speedups on LLaDA-8B-Instruct and Dream-7B-Instruct while keeping task quality.
1 stories, newest first · one card per story
On GSM8K the paper reports 10.12× and 5.12× speedups on LLaDA-8B-Instruct and Dream-7B-Instruct while keeping task quality.
Today 1.5K readers · 2.4K views · 757 reads · all time 189K views