PragWorld：在最小语言改动与会话动态下评估大语言模型局部世界模型的基准
PragWorld: A Benchmark Evaluating LLMs' Local World Model under Minimal Linguistic Alterations and Conversational Dynamics
AI总结：
针对大语言模型在会话中难以维持鲁棒局部世界模型的问题，构建PragWorld基准评估其在最小语言改动下的能力，提出可解释框架定位有害层，并设计层正则化微调策略提升性能。
摘要：
真实世界会话包含大量语用要素，例如实体提及、指称与会话含义。理解这类细微差别是成功实现自然通信的必要条件，通常需要构建能编码上述要素、捕捉其状态演化动态的局部世界模型。然而目前学界尚未明确语言模型（LMs）是否能构建或维持鲁棒的会话隐式表征。本研究评估了语言模型在双人会话中编码、更新内部世界模型的能力，测试其在语言改动下的可塑性。为实现该目标，我们对来自主流数据集的会话施加7种最小语言改动，构建了两个包含是非判断题的基准。我们评估了多款开源与闭源语言模型，发现它们难以维持稳定的准确率。分析显示，语言模型难以记忆关键细节，例如在会话存在语言改动时跟踪实体。随后我们提出双视角可解释性框架，可识别Transformer中有益或有害的层，定位受有害层影响最大的语言改动类型——这类影响通常源于编码虚假信号或依赖捷径。基于上述洞见，我们提出两种基于层正则化的微调策略，用于抑制有害层的影响。
英文摘要：
Real-world conversations are rich with pragmatic elements, such as entity mentions, references, and implicatures. Understanding such nuances is a requirement for successful natural communication, and often requires building a local world model which encodes such elements and captures the dynamics of their evolving states. However, it is not well-understood whether language models (LMs) construct or maintain a robust implicit representation of conversations. In this work, we evaluate the ability of LMs to encode and update their internal world model in dyadic conversations and test their malleability under linguistic alterations. To facilitate this, we apply seven minimal linguistic alterations to conversations sourced from popular datasets and construct two benchmarks comprising yes-no questions. We evaluate a wide range of open and closed source LMs and observe that they struggle to maintain robust accuracy. Our analysis unveils that LMs struggle to memorize crucial details, such as tracking entities under linguistic alterations to conversations. We then propose a dual-perspective interpretability framework which identifies transformer layers that are useful or harmful and highlights linguistic alterations most influenced by harmful layers, typically due to encoding spurious signals or relying on shortcuts. Inspired by these insights, we propose two layer-regularization based fine-tuning strategies that suppress the effect of the harmful layers.
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。
