Stratified Sparse Attention narrows the sparse-to-dense gap
On Ministral-3-8B at 50x sparsity, SSA lifts RULER-HARD from 63.8% to 69.3% at 128K.
50
AI model
Ministral-3-8B
- Mentions · 7d1
- Mentions · 30d1
- All mentions1
- First seenOct 19, 2026
- Last mentioned6d ago
Timeline
1 articles · newest firstAppears with
- FlashInferproduct · 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.