SafeRoute-MAS：多智能体科学推理中基于覆盖与风险的消息路由方法
SafeRoute-MAS: Coverage-Aware and Risk-Calibrated Message Routing for Multi-Agent Scientific Reasoning
一篇关于多智能体消息路由的论文，SafeRoute-MAS 只用集成前信息删消息，能省 12.76% token 还提升 1.80 点成绩。但作者自己用 gpt-oss-120b 做的独立复现基本打平，还给出了诚实的资格判定机制。做多智能体系统、关心评估严谨性的人值得看看这套方法论。
论文提出 SafeRoute-MAS，一种覆盖感知、风险校准的消息路由机制，仅凭集成前信息决定保留或删除多智能体系统中的专家消息。在 5,328 个委员会的密封缓存上，其源兼容运行点带来 +1.80 个百分点增益（95% CI [+1.13, +2.52]），同时删除 12.76% 的消息 token。
作者还在公开的 gpt-oss-120b 端点上对 458 个委员会的分层样本跑了 4,580 次受控集成：原始样本估计为 -0.09 点，按原工作负载加权后为 +0.94 点（95% CI [-1.31, +3.50]），输入 token 减少 18.72%。token 配平的加权对比未证实选择器优势（+1.68 点，CI [-1.27, +5.10]）。88 个缓存非中性标签中仅 13 个（14.8%）在重复调用多数投票下保持方向。
加权后预期腐蚀 UCL 为 4.60%，但 11 个预设边际中只有 2 个通过支持/风险规则。作者据此提出带 Keep-All 回退的数值资格判定机制，并说明该机制在此为回溯性应用，尚未经过前瞻性验证。
SafeRoute-MAS: Coverage-Aware and Risk-Calibrated Message Routing for Multi-Agent Scientific Reasoning
We introduce SafeRoute-MAS, a coverage-aware, risk-calibrated procedure that keeps all specialist messages or removes one using only pre-integration information. On a sealed cache of 5,328 committees, its source-compatible operating point gains +1.80 percentage points (95% CI [+1.13, +2.52]) while removing 12.76% of message tokens. We also run 4,580 controlled integrations over a stratified 458-committee sample on a public gpt-oss-120b endpoint. The sample estimate is -0.09 points; design weighting to the original workload gives +0.94 points (95% CI [-1.31, +3.50]) and 18.72% acted-input reduction. A weighted token-matched comparison likewise does not establish a selector advantage (+1.68 points, CI [-1.27, +5.10]). Only 13 of 88 cached non-neutral labels (14.8%) retain direction by repeated-call majority. Expected-corruption UCL is 4.60% after weighting, but only 2 of 11 prespecified marginals pass the support/risk rule. We therefore propose a numerical qualification with Keep-All fallback, applied here retrospectively rather than validated prospectively.
This report is categorized under AITOP's "Papers" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and AI regulation, compliance frameworks, and sovereign compute policy. Key entities and technical topics under continuous tracking include SafeRoute-MAS, gpt-oss-120b, 多智能体, 智能体.
Source attribution: curated from arXiv cs.AI (@Zhiming Xue, Jianan Liu, Weiran Yan, Aicheng Xie, Jing Yang).