BERT-DST: Scalable End-to-End Dialogue State Tracking with Bidirectional Encoder Representations from Transformer
Abstract
An important yet rarely tackled problem in dialogue state tracking (DST) is scalability for dynamic ontology (e.g., movie, restaurant) and unseen slot values. We focus on a specific condition, where the ontology is unknown to the state tracker, but the target slot value (except for none and dontcare), possibly unseen during training, can be found as word segment in the dialogue context. Prior approaches often rely on candidate generation from n-gram enumeration or slot tagger outputs, which can be inefficient or suffer from error propagation. We propose BERT-DST, an end-to-end dialogue state tracker which directly extracts slot values from the dialogue context. We use BERT as dialogue context encoder whose contextualized language representations are suitable for scalable DST to identify slot values from their semantic context. Furthermore, we employ encoder parameter sharing across all slots with two advantages: (1) Number of parameters does not grow linearly with the ontology. (2) Language representation knowledge can be transferred among slots. Empirical evaluation shows BERT-DST with cross-slot parameter sharing outperforms prior work on the benchmark scalable DST datasets Sim-M
中文速览
对话状态追踪（Dialogue State Tracking, DST）在面对动态本体（如电影、餐厅）和训练时未见过的槽值时，扩展性极差——现有方法要么靠穷举n-gram候选列表效率低下，要么依赖独立的语言理解模块导致错误传播。BERT-DST直接从对话上下文中抽取槽值文本片段，把这个问题转化为类似阅读理解的"找答案"任务：用BERT对对话做上下文编码，再通过分类头判断槽值是"无/不在乎/文本片段"，并用区间预测头定位具体片段，同时让所有槽共享BERT编码器参数以压缩模型规模、促进跨槽知识迁移。实验表明，BERT-DST在专为可扩展DST设计的Sim-M和Sim-R基准上分别达到80.1%和89.6%的联合目标准确率，超越此前最优方法，在标准数据集DSTC2和WOZ 2.0上也取得有竞争力的成绩。这项工作的意义在于，它无需预定义本体或候选生成模块，便能可靠识别未见过的槽值，为面向真实动态场景的对话系统提供了更实用的解决思路。
原文 arXiv:1907.03040;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1907.03040v1