arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2604.14525cs.AI
收藏

量化多查询LLM推理中的跨查询矛盾

Quantifying Cross-Query Contradictions in Multi-Query LLM Reasoning

Virginia Tech(弗吉尼亚理工大学);Columbia University(哥伦比亚大学)

Rohit Kumar Salla, Ramya Manasa Amancherla, Manoj Saravanan

AI总结:

本文研究多查询推理中的逻辑一致性,提出包含390个实例的基准测试,引入集级指标,通过提取承诺、验证全局可满足性及反例引导修复,减少跨查询矛盾并保持单查询准确性。

摘要:

大型语言模型在处理多个相关查询时经常产生相互矛盾的答案。我们研究了案例文件逻辑一致性:在相互依赖的查询中维持全局可满足的信念状态。我们提出一个包含390个多查询推理实例的基准测试,带有蕴含/矛盾/未知标签,并提出集级指标,包括案例可满足率、矛盾密度和修订成本。我们的求解器增强方法提取承诺、验证全局可满足性并执行反例引导修复。在四个推理领域中,我们的方法显著减少了跨查询矛盾(SetCons: 0.56到0.94),同时保持单查询准确性,证明了全局一致性对于稳健的多查询推理至关重要。

英文摘要:

Large language models frequently produce mutually inconsistent answers when reasoning over multiple related queries. We study case-file logical consistency: maintaining a globally satisfiable belief state across interdependent queries. We introduce a benchmark of 390 multi-query reasoning instances with entailment/contradiction/unknown labels and propose set-level metrics including Case Satisfiability Rate, Contradiction Density and Revision Cost. Our solver-augmented approach extracts commitments, verifies global satisfiability and performs counterexample-guided repair. Across four reasoning domains, our method substantially reduces cross-query contradictions (SetCons: 0.56 to 0.94) while preserving per-query accuracy, demonstrating that global coherence is critical for robust multi-query 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。

↑