大模型在主观任务中，思维链提示无法真正推理，反而依赖固定先验。
Larger Language Models Don't Care How You Think: Why Chain-of-Thought Prompting Fails in Subjective Tasks
- 通过思维链提示显式引导推理过程
- 大模型仍受固定先验影响，后验分布几乎不变化
- 适合研究提示工程局限性或主观任务评估的读者
大型语言模型（LLM）的上下文学习（ICL）已成为执行自然语言任务的主要方法，无需基于梯度的参数更新。ICL承诺以极低计算成本实现竞争力甚至顶尖性能。通过在提示中显式包含推理过程，可增强ICL，称为思维链（CoT）提示。然而，近期研究表明，ICL主要依赖任务先验的检索，而非真正“学习”执行任务，尤其在情感、道德等复杂主观领域，先验会固化后验预测。本文研究了‘启用’推理是否也会导致相同行为：即CoT格式是否会检索不变的推理先验，尽管提示中存在证据。结果发现，令人意外的是，对于更大语言模型，CoT同样出现与ICL相同的后验崩溃现象。代码已公开于 https://github.com/gchochla/cot-priors。
原文摘要 · Abstract (English)
In-Context Learning (ICL) in Large Language Models (LLM) has emerged as the dominant technique for performing natural language tasks, as it does not require updating the model parameters with gradient-based methods. ICL promises to "adapt" the LLM to perform the present task at a competitive or state-of-the-art level at a fraction of the computational cost. ICL can be augmented by incorporating the reasoning process to arrive at the final label explicitly in the prompt, a technique called Chain-of-Thought (CoT) prompting. However, recent work has found that ICL relies mostly on the retrieval of task priors and less so on "learning" to perform tasks, especially for complex subjective domains like emotion and morality, where priors ossify posterior predictions. In this work, we examine whether "enabling" reasoning also creates the same behavior in LLMs, wherein the format of CoT retrieves reasoning priors that remain relatively unchanged despite the evidence in the prompt. We find that, surprisingly, CoT indeed suffers from the same posterior collapse as ICL for larger language models. Code is avalaible at https://github.com/gchochla/cot-priors.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。