论文标题
适配器机器人：多合一可控的对话模型
The Adapter-Bot: All-In-One Controllable Conversational Model
论文作者
论文摘要
通过在大型对话数据集中训练大型语言模型来产生连贯和流利的响应的对话模型已经取得了长足的进步。这些模型几乎无法控制产生的响应，并且错过了两个重要特征：连续对话技能集成和无缝利用多样化的知识来源。在本文中，我们提出了适配器 - 机器人，这是一种对话模型，该模型使用了固定的骨干对话模型，例如Dialgpt（Zhang等，2019）和触发器按需对话技能（例如，强烈的响应，天气信息，电影推荐）通过不同的适配器（Houlsby等人，2019年）。每个适配器都可以独立培训，从而可以在不重新训练整个模型的情况下连续整合技能。根据技能，该模型能够以无缝的方式处理多种知识类型，例如文本，表格和图形。对话技能可以通过对话管理器自动触发，也可以手动触发，从而可以高级控制生成的响应。在当前阶段，我们已经实施了12种响应样式（例如，正面，负面等），8种面向目标的技能（例如天气信息，电影推荐等）以及个性化和强调的响应。我们通过将自动评估与现有的最新对话模型进行比较来评估我们的模型，并在apapter.bot.ust.hk上发布了交互式系统。
Considerable progress has been made towards conversational models that generate coherent and fluent responses by training large language models on large dialogue datasets. These models have little or no control of the generated responses and miss two important features: continuous dialogue skills integration and seamlessly leveraging diverse knowledge sources. In this paper, we propose the Adapter-Bot, a dialogue model that uses a fixed backbone conversational model such as DialGPT (Zhang et al., 2019) and triggers on-demand dialogue skills (e.g., emphatic response, weather information, movie recommendation) via different adapters (Houlsby et al., 2019). Each adapter can be trained independently, thus allowing a continual integration of skills without retraining the entire model. Depending on the skills, the model is able to process multiple knowledge types, such as text, tables, and graphs, in a seamless manner. The dialogue skills can be triggered automatically via a dialogue manager, or manually, thus allowing high-level control of the generated responses. At the current stage, we have implemented 12 response styles (e.g., positive, negative etc.), 8 goal-oriented skills (e.g. weather information, movie recommendation, etc.), and personalized and emphatic responses. We evaluate our model using automatic evaluation by comparing it with existing state-of-the-art conversational models, and we have released an interactive system at adapter.bot.ust.hk.
