用排序模型提升大模型推理一致性，更准更快选出正确答案
Boosting Self-Consistency with Ranking
- 将答案选择改造成排序问题，用五个特征综合评分
- 在多个数据集上准确率更高，相同计算量下效果更好
- 适合需要高精度推理的问答任务，尤其擅长复杂题
自一致性通过采样多条推理路径并选择最频繁的答案来提升大语言模型性能，但多数投票法常遗漏已存在于样本中的正确答案。本文提出排名增强型自一致性（RISC），将答案选择重构为排序问题。RISC采用轻量级LambdaRank模型，基于五个精心设计的特征对候选答案进行评分，包括答案频率、语义中心性及推理轨迹一致性。我们在三种数据集上、多种测试时预算条件下评估RISC。结果表明，相比标准自一致性和强基线，RISC在各数据集上均实现更优的准确率-效率权衡，尤其在问答基准上提升显著。进一步分析显示，各特征单独有效且相互补充，凸显了学习融合多种信息信号在测试时答案选择中的价值。
原文摘要 · Abstract (English)
Self-consistency improves large language models by sampling multiple reasoning paths and selecting the most frequent answer, but majority voting often fails to recover correct answers that are already present among the samples. We address this limitation with Ranking-Improved Self-Consistency (RISC), which reformulates answer selection in self-consistency as a ranking problem. Instead of relying on a single uncertainty or confidence signal, RISC uses a lightweight LambdaRank model to score candidate answers with five carefully designed features that capture answer frequency, semantic centrality, and reasoning-trace consistency. We evaluate RISC on three datasets under a range of test-time budgets. Across datasets, RISC consistently achieves a better accuracy-efficiency trade-off than standard self-consistency and strong baselines, with particularly large gains on question answering benchmarks. Further analysis shows that the proposed features are individually useful and, more importantly, complementary, highlighting the value of learning to combine multiple informative signals for test-time answer selection.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。