Meaningless is better: hashing bias-inducing words in LLM prompts improves performance in logical reasoning and statistical learning

TL;DR

This paper presents a novel hashing technique that masks bias-inducing words in LLM prompts, significantly reducing cognitive biases and improving performance across reasoning and statistical tasks, with varying effects on hallucinations.

Contribution

The paper introduces a new hashing method for masking bias-inducing words in LLM prompts, demonstrating its effectiveness across multiple models and tasks.

Findings

Hashing reduces fallacy rates in cognitive bias tests.

Hashing improves task performance in statistical learning.

Effectiveness varies across models and tasks.

Abstract

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.