WISV: Wireless-Informed Semantic Verification for Distributed Speculative Decoding in Device-Edge LLM Inference

arXiv:2604.17701v1 Announce Type: cross Abstract: 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. In this paper, we propose WISV (Wireless-Informed Semantic Verificatio