AI RESEARCH
WISV: Wireless-Informed Semantic Verification for Distributed Speculative Decoding in Device-Edge LLM Inference
arXiv CS.AI
•
ArXi:2604.17701v1 Announce Type: cross While distributed device-edge speculative decoding enhances resource utilization across heterogeneous nodes, its performance is often bottlenecked by conventional token-level verification strategies. Such rigid alignment leads to excessive rejections, significantly diminishing the accepted sequence length and increasing interaction rounds under fluctuating wireless conditions.