Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
AcceptMoE: Commitment-Weighted Self-Sizing Verifier Expert Sets for Efficient MoE Speculative Decoding
Shuang Liang, Hao Mark Chen, Zhiwen Mo +4
Speculative decoding verifies a tree of draft tokens in one target-model forward pass. For a mixture-of-experts (MoE) target, however, parallel verification can activate the union…
cs.LG2025
FastTTS: Accelerating Test-Time Scaling for Edge LLM Reasoning
Hao Mark Chen, Zhiwen Mo, Guanxi Lu +4
Recent advances in reasoning Large Language Models (LLMs) are driving the emergence of agentic AI systems. Edge deployment of LLM agents near end users is increasingly necessary to…