Route-Verify-Vote：面向混合域推理的过程条件自一致性
原标题：Route-Verify-Vote: Procedure-Conditioned Self-Consistency for Mixed-Domain Reasoning
AI 摘要
该论文提出 Route–Verify–Vote（RVV）框架，用于混合域推理中的过程条件自一致性，无需更新模型参数。RVV 通过域标签选择推理过程、逐选项验证约束，并对完整答案集投票，同时将额外采样分配给票数差距小的问题。在 SCoRE 2026 官方测试集上，16 次采样投票达到 74.6% 精确集合准确率，自适应 RVV 达 77.3%，结合模型后达 79.4%，最终系统排名第二。
正文节选
Route–Verify–Vote: Procedure-Conditioned Self-Consistency for Mixed-Domain Reasoning Abstract Compositional generalization remains challenging when language models must combine familiar reasoning operations in unfamiliar ways. The Scenario-Based Commonsense Reasoning Evaluation (SCoRE) 2026 tests this ability on three mixed domains absent from training and requires models to identify the complete set of correct options for each question. We introduce Route–Verify–Vote (RVV), a framework for proc