基于非对称对话中的根本性误解：面向视角主义的MapTask标注方案

摘要

協作對話依賴參與者逐步建立共同基礎，但在非對稱情境中，雙方可能自以為達成共識，實際指涉的卻是不同實體。我們為HCRC地圖任務語料庫（Anderson等人，1991）引入視角主義標註框架，分別捕捉每個指稱表達中說話者與受話者的落地詮釋，從而追蹤理解如何隨時間推移形成、分化與修復。通過採用框架約束的大語言模型標註流程，我們獲得1.3萬個帶可信度評估的指稱表達標註，並分析由此產生的理解狀態。結果表明：當詞彙變體統一後，完全誤解較為罕見，但多重性差異會系統性引發理解分歧，揭示表面共識可能掩蓋指稱錯位。本框架既為研究落地性誤解提供資源與分析視角，也為評估（視覺）大語言模型在協作對話中建模視角依賴性落地的能力奠定基礎。

English

Collaborative dialogue relies on participants incrementally establishing common ground, yet in asymmetric settings they may believe they agree while referring to different entities. We introduce a perspectivist annotation scheme for the HCRC MapTask corpus (Anderson et al., 1991) that separately captures speaker and addressee grounded interpretations for each reference expression, enabling us to trace how understanding emerges, diverges, and repairs over time. Using a scheme-constrained LLM annotation pipeline, we obtain 13k annotated reference expressions with reliability estimates and analyze the resulting understanding states. The results show that full misunderstandings are rare once lexical variants are unified, but multiplicity discrepancies systematically induce divergences, revealing how apparent grounding can mask referential misalignment. Our framework provides both a resource and an analytic lens for studying grounded misunderstanding and for evaluating (V)LLMs' capacity to model perspective-dependent grounding in collaborative dialogue.

基于非对称对话中的根本性误解：面向视角主义的MapTask标注方案

Grounded Misunderstandings in Asymmetric Dialogue: A Perspectivist Annotation Scheme for MapTask

摘要

Support