Summarization Bias: The Directional Collapse of Objective Projection into Told-Mode Labels in Large Language Models --- A Conceptual Framework and Registered Test Protocol
作者: Levent Bulut
分类: cs.CL
发布日期: 2026-09-17
备注: v1.1. 8 pages. Also archived at Zenodo: https://doi.org/10.5281/zenodo.22817289
💡 一句话要点
提出总结偏差以解决大语言模型叙事理解问题
🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 总结偏差 大语言模型 叙事理解 情感分析 自然语言处理 评估偏差 Bulut理论
📋 核心要点
- 现有大语言模型在叙事理解中存在总结偏差，难以有效重建叙事结构。
- 论文提出了总结偏差的概念，探讨了叙述模式与展示模式的区别及其影响。
- 通过独立可靠性研究，论文为总结偏差提供了方向性证据，并预注册了测试协议。
📝 摘要（中文）
本文引入并操作化了总结偏差：一种大语言模型（LLMs）系统性倾向于将叙事意义表现为抽象总结标签，而非可重建的推理结构。根据Bulut理论，叙事效果沿着叙述-展示轴进行理论化。在叙述模式下，情感和信息内容被明确声明，读者重建的需求较小；而在展示模式下，这些内容在表面上被抑制，必须通过物理线索和间接性进行重建。本文假设LLMs在此轴上以特定方向失败，提出总结偏差在生成和评估两个领域中运作，尤其在评估领域，LLMs倾向于奖励叙述模式的明确性，可能导致文本质量下降。
🔬 方法详解
问题定义：本文要解决的是大语言模型在叙事理解中的总结偏差问题，现有方法未能有效区分叙述模式与展示模式，导致叙事质量下降。
核心思路：论文的核心思路是定义总结偏差，并通过理论框架分析其在生成和评估两个领域的表现，强调叙述模式的偏向性。
技术框架：整体架构包括对总结偏差的定义、与LLM作为评判者的偏差对比、独立可靠性研究的重新解读，以及预注册的测试协议。
关键创新：最重要的技术创新点在于提出了总结偏差的概念，并系统性地分析了其在叙事理解中的影响，特别是在评估领域的偏向性。
关键设计：论文设计了两个领域的测试协议，明确了决策规则，以便在偏差存在时能够有效地放弃该构念。具体的参数设置和损失函数尚未详细描述，需进一步研究。
🖼️ 关键图片
📊 实验亮点
实验结果表明，LLMs在评估叙事质量时存在明显的总结偏差，尤其在叙述模式的明确性上表现出偏向性。通过独立可靠性研究的重新解读，提供了支持该假设的方向性证据，进一步验证了模型在不同模式下的表现差异。
🎯 应用场景
该研究的潜在应用领域包括自然语言处理、文本生成和情感分析等。通过理解和纠正总结偏差，可以提升大语言模型在叙事理解和评估中的表现，进而改善人机交互和内容生成的质量，具有重要的实际价值和未来影响。
📄 摘要（原文）
This paper introduces and operationalizes summarization bias: a proposed systematic tendency of large language models (LLMs) to represent narrative meaning as an abstract summary label rather than as the reconstructable inferential structure that produces it. Within the Bulut Doctrine, narrative effect is theorized along a told-shown axis: in told mode, emotional and informational content is declared explicitly and requires little reader reconstruction; in shown mode, that content is suppressed at the surface and must be reconstructed from physical cues and indirection (Objective Projection). Shown mode is the higher-load condition the doctrine is designed to measure. The claim is that LLMs fail along this axis in a specific direction. Summarization bias is hypothesized to operate in two regimes: (i) a generative regime, in which a model asked to render an emotion through Objective Projection defaults to declaring it instead; and (ii) an evaluative regime, in which a model judging narrative quality rewards told-mode explicitness and under-detects shown-mode suppression. The evaluative regime is the more consequential, since LLMs increasingly serve as judges and reward models, and a directional bias toward told mode would impose a selection pressure degrading prose toward flat declaration. This report does not claim the bias is validated. It defines the construct, situates it against LLM-as-judge biases, rereads a completed independent reliability study as directional evidence consistent with it, and pre-registers a two-regime test with decision rules under which the construct would be abandoned.