用前缀一致性提升大模型推理可靠性，节省大量计算资源。
Reliable Chain-of-Thought via Prefix Consistency
- 通过重生成部分推理链，判断答案重现频率来评估可靠性。
- 在多数场景下准确率超越传统多数投票，节省4.6倍至21倍推理令牌。
- 无需概率或评分提示，适用于多种模型和数学科学任务。
大型语言模型在推理任务中常通过采样多个思维链（CoT）并以多数投票（MV）聚合来提升准确率，该方法称为自一致性。我们发现：当截断思维链并重新生成剩余部分时，正确答案的推理路径比错误答案更倾向于重现原答案。基于这一差异，提出前缀一致性作为可靠性信号，按候选答案在重生成中出现的频率加权。该方法无需访问词元概率或自评提示。在五种推理模型和四个数学、科学基准上，前缀一致性在多数设置下是最佳正确性预测器；使用它重加权投票可在最多减少21倍令牌数（中位数4.6倍）的情况下达到标准多数投票的准确率。代码已开源：https://github.com/naoto-iwase/prefix-consistency。
原文摘要 · Abstract (English)
Large Language Models often improve accuracy on reasoning tasks by sampling multiple Chain-of-Thought (CoT) traces and aggregating them with majority voting (MV), a test-time technique called self-consistency. When we truncate a CoT partway through and regenerate the remainder, we observe that traces with correct answers reproduce their original answer more often than traces with wrong answers. We use this difference as a reliability signal, prefix consistency, that weights each candidate answer by how often it reappears under regeneration. It requires no access to token log-probabilities or self-rating prompts. Across five reasoning models and four math and science benchmarks, prefix consistency is the best correctness predictor in most settings, and reweighting votes by it reaches Standard MV plateau accuracy at up to 21x fewer tokens (median 4.6x). Our code is available at https://github.com/naoto-iwase/prefix-consistency.
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。