无意义即更好：对 LLM 提示词中诱发偏见的词语进行哈希可提升逻辑推理与统计学习性能
Meaningless is better: hashing bias-inducing words in LLM prompts improves performance in logical reasoning and statistical learning
AI总结：
提出用无意义哈希标识符遮蔽 LLM 提示词中诱发偏见的词语，并在 490 条提示词、五类模型上验证其可改善逻辑推理与统计学习，但效果因模型和任务而异。
摘要：
本文介绍了一种名为“hashing（哈希）”的新方法，该方法使用类似哈希的无意义标识符来遮蔽大型语言模型（LLM）提示词中可能诱发偏见的词语，以减少认知偏见以及对外部知识的依赖。该方法在三组实验中接受测试，共涉及 490 条提示词。使用卡方检验进行的统计分析表明，在所有测试场景中均取得显著改进，测试覆盖 LLama、ChatGPT、Copilot、Gemini 和 Mixtral 模型。在第一项实验中，hashing 降低了改进版“Linda”问题中的谬误率，该问题旨在评估模型对认知偏见的易感性。在第二项实验中，它提升了 LLM 在频繁项集提取任务上的结果。在第三项实验中，我们发现当 Linda 问题以表格形式而非文本形式呈现时，hashing 仍然有效，这表明该技术适用于多种输入表示。总体而言，该方法被证明能够改善偏见减少以及外部知识的纳入。尽管偏见有所减少，但不同类型 LLM 模型的幻觉率下降并不一致。这些发现表明，遮蔽诱发偏见的术语可以提升 LLM 性能，但其有效性取决于模型和任务。
英文摘要：
This paper introduces a novel method, referred to as "hashing", which involves masking potentially bias-inducing words in large language models (LLMs) with hash-like meaningless identifiers to reduce cognitive biases and reliance on external knowledge. The method was tested across three sets of experiments involving a total of 490 prompts. Statistical analysis using chi-square tests showed significant improvements in all tested scenarios, which covered LLama, ChatGPT, Copilot, Gemini and Mixtral models. In the first experiment, hashing decreased the fallacy rate in a modified version of the "Linda" problem aimed at evaluating susceptibility to cognitive biases. In the second experiment, it improved LLM results on the frequent itemset extraction task. In the third experiment, we found hashing is also effective when the Linda problem is presented in a tabular format rather than text, indicating that the technique works across various input representations. Overall, the method was shown to improve bias reduction and incorporation of external knowledge. Despite bias reduction, hallucination rates were inconsistently reduced across types of LLM models. These findings suggest that masking bias-inducing terms can improve LLM performance, although its effectiveness is model- and task-dependent.
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。
