GUT：基于图复杂度量化与优化大语言模型的推理不确定性
GUT: Quantifying and Optimizing the Reasoning Uncertainty of LLMs via Graph Complexity
- Nanjing University（南京大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究针对大语言模型推理存在的不确定性问题，提出GUT方法，通过构建图表征推理分支，设置量化与优化模块，在多模型多数据集上验证了方法的有效性。
AI中文摘要：
近年来，大语言模型（LLMs）的推理能力取得了巨大进展。然而，LLMs的推理过程常存在不确定性：即便输入相同提示，LLMs在每一步推理时也会产生大量分歧分支，部分分支的推理链和结果明显不可靠，甚至毫无意义。本文提出基于图复杂度的不确定性方法（GUT），用于研究LLMs的推理不确定性。GUT的核心思路是用有向无环图表征每条推理链的潜在分支，确保图空间内全面覆盖所有潜在分支。基于该认知，我们进一步构建了GUT的两个模块：量化模块（GUT-Q）和优化模块（GUT-O），分别用于量化和降低LLMs的推理不确定性。GUT-Q通过图复杂度近似推理空间复杂度，以此衡量LLMs的推理不确定性；GUT-O将负不确定性作为强化学习中的奖励函数，实现不确定性优化。在四个LLMs和五个数据集上开展的实验验证了GUT的有效性。
英文摘要：
Recent years have witnessed great advances in the reasoning ability of Large Language Models (LLMs). However, the reasoning processes of LLMs often exhibit uncertainty, where LLMs often produce a proliferation of divergent branches at each reasoning step even when fed the same prompting inputs, and certain branches exhibit evidently incredible, even nonsensical, reasoning chains and results. In this paper, we propose the Graph-complexity-based UncerTainty (GUT) method for investigating the reasoning uncertainty of LLMs. The key idea of GUT is to characterize the potential branches of each reasoning chain with a directed acyclic graph, thereby ensuring that all potential branches are comprehensively covered within the graph space. Building upon this recognition, we further build two modules of GUT, that is, a Quantification (GUT-Q) module and an Optimization (GUT-O) module, for quantifying and reducing the reasoning uncertainty of LLMs, respectively. GUT-Q measures LLM reasoning uncertainty by approximating the reasoning space complexity with graph complexity. GUT-O implements uncertainty optimization by treating negative uncertainty as the reward function in reinforcement learning. Experimental results conducted on four LLMs and five datasets validate the effectiveness of GUT.
