对于 GPT-4 而言与人类相同：信息结构预测长距离依存关系的可接受性
For GPT-4 as with Humans: Information Structure Predicts Acceptability of Long-Distance Dependencies
AI总结：
本文通过人类同款零样本任务探测 GPT-4，发现其能复现信息结构对长距离依存关系可接受性的预测作用及因果效应。
摘要：
任何语言模型（LM）在多大程度上理解自然语言或生成可靠的元语言判断，仍存在争议。此外，相对较少的研究证明 LM 能够表征并遵循语言学家提出的形式与功能之间的微妙关系。我们在此聚焦近期研究确立的一种特定关系：在广泛的基础构式和多种类型的长距离依存关系（LDD）中，英语使用者对规范句信息结构的判断，可预测独立收集的相应“长距离依存关系”构式可接受性评分。为确定是否有 LM 捕捉到这一关系，我们在用于人类的相同任务及新的扩展任务上探测 GPT-4。结果显示，尽管任务具有零样本、显式的性质，且几乎不存在污染可能，GPT-4 在信息结构和可接受性任务上表现出可靠的元语言技能，并复现了两者之间显著的交互作用（研究 1a、1b）。研究 2 操纵基础句的信息结构并证实因果关系：提高上下文句中某一成分的显著性，会提升随后对 LDD 构式的可接受性评分。研究结果表明，自然英语与 GPT-4 生成的英语之间、以及信息结构与句法之间存在紧密关系，值得进一步探索。
英文摘要：
It remains debated how well any LM understands natural language or generates reliable metalinguistic judgments. Moreover, relatively little work has demonstrated that LMs can represent and respect subtle relationships between form and function proposed by linguists. We here focus on a particular such relationship established in recent work: English speakers' judgments about the information structure of canonical sentences predicts independently collected acceptability ratings on corresponding 'long distance dependency' [LDD] constructions, across a wide array of base constructions and multiple types of LDDs. To determine whether any LM captures this relationship, we probe GPT-4 on the same tasks used with humans and new extensions.Results reveal reliable metalinguistic skill on the information structure and acceptability tasks, replicating a striking interaction between the two, despite the zero-shot, explicit nature of the tasks, and little to no chance of contamination [Studies 1a, 1b]. Study 2 manipulates the information structure of base sentences and confirms a causal relationship: increasing the prominence of a constituent in a context sentence increases the subsequent acceptability ratings on an LDD construction. The findings suggest a tight relationship between natural and GPT-4 generated English, and between information structure and syntax, which begs for further exploration.
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。
