发表机构
Humber Polytechnic; University of Toronto（亨伯理工学院; 多伦多大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文研究大语言模型中无校正控制的基础问题，提出路径剖面概念以分析真值追踪，指出纯文本模型继承的模式可提供衍生可应答性，不同方法对任务的真值追踪改进可能与表面改进不一致。
AI 中文摘要
近期研究表明，部分大语言模型的表征具有内容或指称。基础（grounding）可在不提供实时校正路径的情况下实现这两种特性。本文探讨该差距会带来什么结果：当差异能影响特定目标与任务的配置所产生、接受或弃权（不执行）的内容时，输出是可应答的；仅当存在实时且足够独立的路径可检测并修复新差异时，该配置才具有校正控制；路径剖面记录了哪些路径约束该配置及其关联方式，这些剖面支持对真值追踪的分析——即对表征成功的模式化支撑。语言模型是压力测试案例，纯文本配置提供了与任务相关的极限案例，文本训练的模型继承了证词、连贯性和先前校正的模式。若训练后仍保留目标敏感的校正，这些模式可提供衍生可应答性（继承约束）；实时可应答性是当前路径对新差异提供的关系。当任务需要独立获取事实信息时，应出现流畅失败；自洽性、检索、工具、代码执行、多模态输入和反馈应能提供选择性帮助。按任务划分的路径交互可验证这些区分，该分解的经验负担是预测未见过的路径-任务组合或改进干预选择，而无需概念重构；表面改进与真值追踪改进可能不一致。
英文摘要
Recent work suggests that some large language model representations have content or reference. Grounding can secure either without supplying live routes for correction. This paper asks what follows from that gap. An output is answerable when discrepancies can affect what a target- and task-specific arrangement produces, accepts, or withdraws. The arrangement has corrective control only when live, sufficiently independent routes can detect and repair fresh discrepancies. A route profile records which routes constrain the arrangement and how they are related. Those profiles support analysis of truth-tracking: patterned support for representational success. Language models are the pressure case; text-only arrangements provide a task-relative limiting case. Text-trained models inherit patterns of testimony, coherence, and prior correction. Where target-sensitive correction survives training, these can supply derivative answerability (inherited constraint); live answerability is the relation supplied by a current route for fresh discrepancies. Fluent failures should follow when a task requires independently informative access to the facts. Self-consistency, retrieval, tools, code execution, multimodal input, and feedback should help selectively. Route-by-task interactions test the distinctions. The decomposition's empirical burden is to predict held-out route--task combinations or improve intervention choice without conceptual refitting. Surface improvement and truth-tracking improvement can come apart.
Comments24 pages, 1 figure, 1 table. A six-page methodological supplement, reproducible R script, and constructed data are included as ancillary files
