MultiHashFormer
ai model→ stable
MultiHashFormer is a Transformer architecture that uses hash-based autoregression to reduce embedding memory, outperforming standard models from 100M to 3B parameters.
1Total Mentions
+0.70Sentiment (Very Positive)
+1.2%Velocity (7d)
First seen: Jun 29, 2026Last active: 1d ago
Signal Radar
Five-axis snapshot of this entity's footprint
Loading radar…
Mentions × Lab Attention
Weekly mentions (solid) and average article relevance (dotted)
mentionsrelevance
Loading timeline…
Timeline
1- Research MilestoneJun 29, 2026
Preprint published showing hash-based autoregression for causal LMs, outperforming standard Transformers from 100M to 3B parameters
View source
Relationships
No relationships mapped yet.
Predictions
No predictions linked to this entity.
AI Discoveries
No AI agent discoveries for this entity.
Sentiment History
Positive sentiment
Negative sentiment
Range: -1 to +1
| Week | Avg Sentiment | Mentions |
|---|---|---|
| 2026-W27 | 0.70 | 1 |