Frequency-aware remasking lifts diffusion language model test-time scaling
MATH-500 gains reach 3.73 points for LLaDA-8B-Instruct and 7.33 points for Dream-7B-Instruct.
34
AI model
Dream-7B-Instruct
- Mentions · 7d1
- Mentions · 30d5
- All mentions5
- First seenOct 21, 2026
- Last mentioned5d ago
Timeline
5 articles · newest firstdSpark 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
Learned Unmasking Policies Improve Diffusion Language Model Efficiency
A 856K-parameter policy trained on trajectory-contraction targets achieves best accuracy-NFE frontier on LLaDA-8B-Instruct.
45
Test-time Regression Improves Sequence Compression
TTR-Compressor uses end-to-end learned regression for sequence compression, outperforming pooling and projections in language and multimodal models.
40
Study isolates what improves learned unmasking in diffusion language models
An 856K-parameter policy on LLaDA-8B-Instruct leads the accuracy–NFE frontier on GSM8K and largely transfers to Dream-7B-Instruct.
36
Appears with
- LLaDA-8B-Instructai model · 5
- VILA-Laborganization · 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.