ACED:用于数学推理的自然语言与可执行证据的参考锚定聚合
ACED: Reference-Anchored Aggregation of Natural-Language and Executable Evidence for Mathematical Reasoning
AI总结:
ACED是参考锚定框架,将自然语言自洽性答案设为默认、可执行输出为支持证据,经实验在五个数学应用题基准上表现优于或匹配基线,可可靠整合两类推理证据。
摘要:
自然语言自洽性与可执行程序为数学推理提供了互补的测试时间证据,但任一渠道都可能存在相同的语义错误。因此我们提出ACED,这是一个参考锚定框架,将自洽性答案视为默认结果,将可执行输出视为支持证据而非可互换的投票。ACED将特定来源的答案计数转换为平滑支持度,通过加权对数池对其进行组合,且仅在校准后的相对证据余量被超过时才会覆盖参考结果。该规则同时支持候选选择与仅可执行答案的接纳。在五个数学应用题基准上,ACED的表现与对应基线相当或更优。在GSM8K上,ACED-CANDIDATE仅通过三次破坏性覆盖就恢复了可达候选集增益的89.7%,而ACED-ALL达到86.9%的准确率,相比之下,直接程序思维聚合的准确率为82.6%。这些结果表明,保留证据来源能够实现更可靠的自然语言与可执行推理的整合。
英文摘要:
Natural-language self-consistency and executable programs provide complementary test-time evidence for mathematical reasoning, yet either channel may support the same semantic error. Thus we propose ACED, a reference-anchored framework that treats the self-consistency answer as the default and executable outputs as supporting evidence rather than interchangeable votes. ACED converts source-specific answer counts into smoothed supports, combines them through a weighted logarithmic pool, and overrides the reference only when a calibrated relative-evidence margin is exceeded. The same rule supports both candidate selection and admission of executable-only answers. Across five math word-problem benchmarks, ACED matches or outperforms the corresponding baselines. On GSM8K, ACED-CANDIDATE recovers 89.7% of the attainable candidate-set gain with only three damaging overrides, while ACED-ALL reaches 86.9% accuracy versus 82.6% for direct Program-of-Thought aggregation. These results show that preserving evidence provenance enables more reliable integration of natural-language and executable reasoning.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。