用胶囊网络捕捉语义角色标注中的论元交互
Capturing Argument Interaction in Semantic Role Labeling with Capsule Networks
摘要：
语义角色标注（SRL）涉及从自然语言句子中提取命题（即谓词及其类型化参数）。最先进的SRL模型依赖于强大的编码器（例如，LSTM），并且不对参数之间的非局部交互进行建模。我们提出了一种新的方法来建模这些相互作用，同时保持有效的推理。具体来说，我们使用胶囊网络：每个命题被编码为\textit{capsules}的元组，每个参数类型（即角色）一个胶囊。这些元组充当整个命题的嵌入。在每一个网络层中，胶囊彼此交互，并与句子中的单词表示交互。每次迭代都会导致更新的命题嵌入和关于SRL结构的更新预测。我们的模型在所有7种CoNLL-2019语言上的表现都大大优于非细化基线模型，并在5种语言（包括英语）上实现了最先进的依赖SRL结果。我们分析的类型的错误纠正的细化程序。例如，每个角色通常（但不总是）最多填充一个参数。然而，强制执行这种近似约束是没有用的现代SRL系统，迭代过程纠正错误，捕捉这种直觉在一个灵活的和上下文敏感的方式。
英文摘要：
Semantic role labeling (SRL) involves extracting propositions (i.e. predicates and their typed arguments) from natural language sentences. State-of-the-art SRL models rely on powerful encoders (e.g., LSTMs) and do not model non-local interaction between arguments. We propose a new approach to modeling these interactions while maintaining efficient inference. Specifically, we use Capsule Networks: each proposition is encoded as a tuple of \textit{capsules}, one capsule per argument type (i.e. role). These tuples serve as embeddings of entire propositions. In every network layer, the capsules interact with each other and with representations of words in the sentence. Each iteration results in updated proposition embeddings and updated predictions about the SRL structure. Our model substantially outperforms the non-refinement baseline model on all 7 CoNLL-2019 languages and achieves state-of-the-art results on 5 languages (including English) for dependency SRL. We analyze the types of mistakes corrected by the refinement procedure. For example, each role is typically (but not always) filled with at most one argument. Whereas enforcing this approximate constraint is not useful with the modern SRL system, iterative procedure corrects the mistakes by capturing this intuition in a flexible and context-sensitive way.
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。
