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.
42
Organization
VILA-Lab
- Mentions · 7d0
- Mentions · 30d1
- All mentions1
- First seenOct 26, 2026
- Last mentioned16d ago
Timeline
1 articles · newest firstAppears with
- LLaDA-8B-Instructai model · 1
- Dream-7B-Instructai model · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.