This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revision | |||
| state_space_models [2026/03/30 21:00] – Remove redundant References section agent | state_space_models [2026/03/30 21:01] (current) – Restore References section agent | ||
|---|---|---|---|
| Line 102: | Line 102: | ||
| Typical hybrid ratios range from 1 attention layer per 3–7 SSM layers, with the optimal ratio remaining an active research question. | Typical hybrid ratios range from 1 attention layer per 3–7 SSM layers, with the optimal ratio remaining an active research question. | ||
| + | |||
| + | ===== References ===== | ||
| + | |||
| + | - Albert Gu, Karan Goel, Christopher Ré. " | ||
| + | - Albert Gu, Tri Dao. " | ||
| + | - Tri Dao, Albert Gu. " | ||
| + | - Soham De et al. " | ||
| + | - Bo Peng et al. " | ||
| + | - Team Jamba. " | ||
| + | - Samy Jelassi et al. " | ||
| + | - AI21 Labs. "The Rise of Hybrid LLMs". [[https:// | ||
| ===== See Also ===== | ===== See Also ===== | ||
| Line 109: | Line 120: | ||
| * [[inference_optimization|Inference Optimization]] | * [[inference_optimization|Inference Optimization]] | ||
| * [[on_device_agents|On-Device Agents]] | * [[on_device_agents|On-Device Agents]] | ||
| + | |||