New AlterNet architecture combines mixture-of-experts with learned halting for elastic LLM inference
Researchers propose AlterNet, an LLM architecture pairing adaptive-depth mixture-of-experts with learned halting, allowing compute demands to adjust at inference rather than being fixed at training time.
Read the full story at arXiv cs.AI ↗
Timeline · 1 report
- 2026-11-09 05:00 · arXiv cs.AI
AlterNet: Adaptive-Depth Mixture-of-Experts with Learned Halting for Elastic Inference