Scholarlink

Large Language Models (LLMs) have made significant advancements in addressing diverse natural language processing tasks. However, their performance is often limited by inherent comprehension of problems. To address this limitation, we propose Exchange-of-Perspective (EoP), a novel framework designed to exchange perspectives across different definitions of problem, so that it can break the fixed mindset from any particular formulation of the question. We conducted extensive and comprehensive experiments on 8 benchmarks. The results show that EoP can significantly improve performance. For instance, compared to the non-commutative baseline PHP, with GPT-3.5-Turbo and EoP, we observe a +3.6% improvement on AQuA (60.6% \(\rightarrow \) 64.2%), while GPT-4-powered EoP demonstrates a +7.7% enhancement on Math (53.9% \(\rightarrow \) 61.6%) and a +3.5% improvement on OlympiadBench (43.5% \(\rightarrow \) 47.0%) when using Qwen-2.5-72b.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Exchange of Perspective Prompting Enhances Reasoning in Large Language Models

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Lin Sun,
  • Can Zhang

摘要

Large Language Models (LLMs) have made significant advancements in addressing diverse natural language processing tasks. However, their performance is often limited by inherent comprehension of problems. To address this limitation, we propose Exchange-of-Perspective (EoP), a novel framework designed to exchange perspectives across different definitions of problem, so that it can break the fixed mindset from any particular formulation of the question. We conducted extensive and comprehensive experiments on 8 benchmarks. The results show that EoP can significantly improve performance. For instance, compared to the non-commutative baseline PHP, with GPT-3.5-Turbo and EoP, we observe a +3.6% improvement on AQuA (60.6% \(\rightarrow \) 64.2%), while GPT-4-powered EoP demonstrates a +7.7% enhancement on Math (53.9% \(\rightarrow \) 61.6%) and a +3.5% improvement on OlympiadBench (43.5% \(\rightarrow \) 47.0%) when using Qwen-2.5-72b.