1
Sequential Consensus for Multi-Agent LLM Debates: A Wald-SPRT compute governor with calibration-based failure detection
用Wald的序贯概率比检验动态调控LLM辩论轮数，显著提升推理效率与资源利用率
arXiv:2605.19193v1 Announce Type: new Abstract: Multi-agent LLM debate improves factuality and reasoning, but most recipes pick a fixed round count, o…