arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.06145cs.CL
收藏

提示复杂性:大语言模型中用于文本和行为的最短提示

Prompting Complexity: Shortest Prompts for Texts and Behaviors in LLMs

Dalle Molle Institute for Artificial Intelligence (IDSIA)(达勒·莫利人工智能研究所)

Adrian Cosma

AI总结:

研究提示复杂性,定义其为固定语言模型产生目标文本的最短合理提示量,类似资源受限柯尔莫哥洛夫复杂性。介绍计算方法、扩展定义到软提示复杂性,定义提示距离等,据此制定研究议程以探究短提示可实现的文本和行为。

摘要:

在本文中,我们定义了提示复杂性的量:对于一个固定的指令微调语言模型,能使确定性解码产生目标文本的最短合理提示是什么?它是资源受限的柯尔莫哥洛夫复杂性的语言模型相关类似物。与经典柯尔莫哥洛夫复杂性不同,此度量并非通用。在有限上下文设置中可通过枚举计算,但不存在与模型无关的不变性定理。为使搜索空间与提示工程一致,我们将程序限制为合理的人类可读文本。我们将精确定义扩展到近似输出的软提示复杂性,得出模型相关文本压缩的有损概念及提示优化的正式目标。还通过比较最短生成提示定义了提示距离,以及达到满足规范的任何输出的行为提示复杂性。基于这些公式,我们定义了一个研究议程,用于实证研究在固定语言模型接口下,哪些文本和行为可从短的合理提示中获得。

英文摘要:

In this paper, we define the quantity of prompting complexity: for a fixed instruction-tuned language model, what is the shortest plausible prompt that makes deterministic decoding produce a target text? It is an LM-relative analogue of resource-bounded Kolmogorov complexity: the prompt is a program, the model interface is the interpreter, and information omitted from the prompt is supplied by the model's weights, training distribution, tokenizer, template, and decoding rule. Unlike classical Kolmogorov complexity, this measure is intentionally non-universal. In the finite-context setting it is computable by enumeration, but there is no model-independent invariance theorem; the same text may be cheap for one model and inaccessible or expensive for another. To keep the search space aligned with prompt engineering, we restrict programs to plausible human-readable texts rather than arbitrary token strings. We extend the exact definition to soft prompting complexity for approximate outputs, yielding a lossy notion of model-relative text compression and a formal target for prompt optimization. We also define prompting distance by comparing shortest generating prompts, and behavioral prompting complexity for reaching any output satisfying a specification. Based on these formulations, we define a research agenda for empirically studying which texts and behaviors are accessible from short plausible prompts under a fixed LM interface.

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。

↑