Artificial entrepreneurial cognition: Locating and causally steering an opportunity recognition dial inside large language models (LLMs)
作者: Christian Fisch, Angela Altmeier, Martin Obschonka, Michal Kosinski, Pin Ni
分类: cs.CL, cs.LG
发布日期: 2026-09-14
备注: 105 pages, 5 figures, 20 tables (8 in the main text, 12 in the appendices)
💡 一句话要点
提出人工创业认知以探索大语言模型中的机会识别机制
🎯 匹配领域: 支柱一:机器人控制 (Robot Control) 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 人工创业认知 机会识别 大语言模型 因果干预 机制可解释性 创业研究 表征工程
📋 核心要点
- 现有的创业认知研究主要集中在人类行为者,缺乏对大语言模型内部表征的探索。
- 本文提出人工创业认知,通过对大语言模型的机会识别进行因果干预,探索其内部表征。
- 实验结果表明,机会识别方向可被恢复且具有显著影响,且与机会评估和利用方向有明确区分。
📝 摘要(中文)
创业认知是创业研究的基础。然而,随着大语言模型(LLMs)在创业工作中的日益参与,认知问题超越了人类行为者,扩展到内部表征尚未被充分探索的系统。本文引入了人工创业认知,定义了人工智能系统内部与创业相关的表征和计算的功能组织。我们通过表征工程将机制可解释性引入创业研究,专注于机会识别(OR),构建了636对匹配的OR存在和缺失场景,并在Llama 3.1 8B-Instruct中恢复了OR方向。我们直接干预这一方向,调整模型的机会判断。我们的研究首次在LLM内部对创业构念的内部表征进行因果干预,结果显示机会识别与评估之间的区别具有具体的表征形式。
🔬 方法详解
问题定义:本文旨在解决大语言模型内部创业认知的表征问题,现有方法未能有效探索这些内部机制。
核心思路:通过构建机会识别的场景对,直接干预模型的内部表征,调整其机会判断,从而实现对创业认知的因果操控。
技术框架:研究设计包括构建636对场景,使用Llama 3.1 8B-Instruct模型进行干预,评估机会识别的方向及其对模型判断的影响。
关键创新:首次在大语言模型内部对创业构念进行因果干预,明确了机会识别与评估之间的区别,推动了创业理论与AI系统的结合。
关键设计:采用了特定的场景对构建方法,确保OR存在与缺失的匹配,同时使用了行为消融和几何比较等技术手段进行验证。实验涵盖了四个不同规模和类型的LLM,确保结果的普适性。
🖼️ 关键图片
📊 实验亮点
实验结果显示,机会识别方向的干预能够显著影响模型的判断,且与机会评估和利用方向有明确的几何分离。通过多种验证方法,结果的可靠性得到了增强,展示了该方法的有效性和创新性。
🎯 应用场景
该研究为创业领域提供了新的视角,能够帮助创业者和研究者理解大语言模型如何影响机会识别过程。未来,该方法可应用于创业教育、商业决策支持系统等领域,推动AI在创业实践中的应用。
📄 摘要(原文)
Entrepreneurial cognition is a foundation of entrepreneurship research. Yet the growing involvement of large language models (LLMs) in entrepreneurial work extends the cognition question beyond human actors to systems whose internal representations remain largely unexplored. We introduce artificial entrepreneurial cognition, the functional organisation of entrepreneurship-relevant representations and computations inside artificial intelligence (AI) systems. We bring mechanistic interpretability into entrepreneurship research through representation engineering. Focusing on opportunity recognition (OR), we construct 636 matched OR-present and OR-absent scenario pairs and recover an OR direction in Llama 3.1 8B-Instruct. Rather than infer the construct from outputs, we intervene directly on this direction, steering the model up and down along what we call the opportunity recognition dial, and its opportunity judgments shift with it. To our knowledge, this is the first causal intervention on an internal representation of an entrepreneurship construct inside an LLM. Held-out tests, lexical and topical controls, behavioural ablation, and geometric comparisons show that the direction is recoverable, consequential, and distinct from the opportunity evaluation and exploitation directions, although steering it also shifts judgments about these neighbouring stages. Recovery, signed steering, and geometric separation hold across four additional LLMs spanning different scales and families. These results give the contested distinction between opportunity recognition and evaluation a concrete representational form inside AI systems. More broadly, they establish internal representations as a new object of entrepreneurship inquiry and show how entrepreneurship theory can guide their identification, causal manipulation, and interpretation.