最先进的对话模型仍然对事实准确性和自我矛盾甚至困难。轶事，他们已被观察到在整个话语中未能维持性质身份;更具体地，可能会涉及其对话者的作用。在这项工作中，我们正规化和量化这种缺陷，并通过人类评估实验表明这确实是一个问题。相比之下，我们展示了专门识别谁在谈话的歧视模型可以表现良好;此外，这些可以用作自动指标。最后，我们评估了各种缓解方法，包括模型架构，培训协议和解码策略的变化。根据人类的注释者，我们最好的车型减少了近65％的误认为是近65％，同时提高了参与度。尽管有这些结果，但我们发现维持性格身份仍然是一个具有挑战性的问题。

Chit-chat models are known to have several problems: they lack specificity, do not display a consistent personality and are often not very captivating. In this work we present the task of making chit-chat more engaging by conditioning on profile information. We collect data and train models to (i) condition on their given profile information; and (ii) information about the person they are talking to, resulting in improved dialogues, as measured by next utterance prediction. Since (ii) is initially unknown, our model is trained to engage its partner with personal topics, and we show the resulting dialogue can be used to predict profile information about the interlocutors.

In open-domain dialogue intelligent agents should exhibit the use of knowledge, however there are few convincing demonstrations of this to date. The most popular sequence to sequence models typically "generate and hope" generic utterances that can be memorized in the weights of the model when mapping from input utterance(s) to output, rather than employing recalled knowledge as context. Use of knowledge has so far proved difficult, in part because of the lack of a supervised learning benchmark task which exhibits knowledgeable open dialogue with clear grounding. To that end we collect and release a large dataset with conversations directly grounded with knowledge retrieved from Wikipedia. We then design architectures capable of retrieving knowledge, reading and conditioning on it, and finally generating natural responses. Our best performing dialogue models are able to conduct knowledgeable discussions on open-domain topics as evaluated by automatic metrics and human evaluations, while our new benchmark allows for measuring further improvements in this important research direction.

我们提出了Blenderbot 3，这是一个175B参数对话模型，能够通过访问Internet和长期内存进行开放域对话，并接受了大量用户定义的任务的培训。我们同时发布了模型权重和代码，还将模型部署在公共网页上，以与有机用户进行交互。该技术报告描述了该模型的构建方式（建筑，模型和培训计划）以及其部署的细节，包括安全机制。人类评估表明，它优于现有的开放域对话代理，包括其前身（Roller等，2021； Komeili等，2022）。最后，我们使用部署收集的数据详细介绍了持续学习的计划，该数据也将公开发布。因此，该研究计划的目标是使社区能够研究通过互动学习的不断改进的负责任的代理商。

The goal of building dialogue agents that can converse with humans naturally has been a long-standing dream of researchers since the early days of artificial intelligence. The well-known Turing Test proposed to judge the ultimate validity of an artificial intelligence agent on the indistinguishability of its dialogues from humans'. It should come as no surprise that human-level dialogue systems are very challenging to build. But, while early effort on rule-based systems found limited success, the emergence of deep learning enabled great advance on this topic. In this thesis, we focus on methods that address the numerous issues that have been imposing the gap between artificial conversational agents and human-level interlocutors. These methods were proposed and experimented with in ways that were inspired by general state-of-the-art AI methodologies. But they also targeted the characteristics that dialogue systems possess.

当前的语言模型达到了较低的困惑，但其产生的几代人仍然遭受有毒的反应，重复性和矛盾。标准语言建模设置无法解决这些问题。在本文中，我们介绍了一个新的体系结构{\ sc导演}，由一个统一的生成器分类器组成，具有语言建模和每个输出令牌的分类头。培训是使用标准语言建模数据共同进行的，并以所需和不良序列标记的数据。与标准语言模型相比，该模型在多种设置中的实验表明，该模型具有竞争性的培训和解码速度，同时产生了较高的结果，从而减轻了已知的问题，同时保持发电质量。就准确性和效率而言，它还优于现有的模型指导方法。

经过培训的模拟静态数据集的冷冻模型永远无法提高其性能。可以采用互联网进行互联网以获取最新信息并在部署期间从人类那里获得反馈的模型提供了适应新信息并提高其性能的承诺。在这项工作中，我们研究了如何在此类学习框架中提高以互联网为导向的对话技能。我们收集人类互动的部署数据，并公开可用，并收集各种类型的人类反馈 - 包括二进制质量测量，自由形式的文本反馈和罚款良好的失败原因。然后，我们研究了各种从此类反馈中改进的算法，包括标准监督学习，拒绝抽样，模型引导和基于奖励的学习，以便对哪种类型的反馈和算法效果最好。我们发现最近介绍的导演模型（Arora等人，'22）比其他现有方法显示出显着改善。

对于谈话的AI和虚拟助手以现实的方式与人类沟通，他们必须表现出人类特征，例如情感和个性的表达。目前对构建人类对话剂的尝试呈现出显着的困难。我们提出基于Tropes的人为水平属性（HLA）作为学习对话代理的方法，这些方法可以模仿虚构人物的个性。 Tropes是由观察员的次要观察和确定的虚构个性的特征。通过将详细的HLA数据与特定字符的对话数据组合，我们提供了一个数据集，HLA-Chat，模型字符配置文件，并提供对话代理通过HLA学习角色语言样式的能力。然后，我们介绍了一个三组件系统，Aloha（代表人工学习人为学习），它结合了字符空间映射，角色社区检测和语言样式检索，以构建特定字符（或个性）特定语言模型。我们的初步实验表明Aloha的两种变化与我们提出的数据集相结合，可以在识别所选择的目标字符的正确对话响应时占据基线模型，并且无论字符的身份，节目类型如何，都是稳定的对话。

Conversational AI has become an increasingly prominent and practical application of machine learning. However, existing conversational AI techniques still suffer from various limitations. One such limitation is a lack of well-developed methods for incorporating auxiliary information that could help a model understand conversational context better. In this paper, we explore how persona-based information could help improve the quality of response generation in conversations. First, we provide a literature review focusing on the current state-of-the-art methods that utilize persona information. We evaluate two strong baseline methods, the Ranking Profile Memory Network and the Poly-Encoder, on the NeurIPS ConvAI2 benchmark dataset. Our analysis elucidates the importance of incorporating persona information into conversational systems. Additionally, our study highlights several limitations with current state-of-the-art methods and outlines challenges and future research directions for advancing personalized conversational AI technology.

大型语言模型可以产生流畅的对话，但往往是幻觉的事实不准确。虽然检索式增强的模型有助于缓解这个问题，但他们仍然面临着推理的艰难挑战，以便同时提供正确的知识和产生对话。在这项工作中，我们提出了一种模块化模型，知识响应（K2R），将知识纳入会话代理商，这将这个问题分解为两个更简单的步骤。 K2R首先生成一个知识序列，给定对话背景作为中间步骤。在此“推理步骤”之后，该模型随后参加自己生成的知识序列，以及对话背景，以产生最终的响应。在详细的实验中，我们发现这种模型在知识接地的对话任务中少幻觉，并且在可解释性和模块化方面具有优势。特别地，它可以用来将QA和对话系统一起融合在一起，以使对话代理能够提供知识渊博的答案，或者QA模型，以在零拍摄设置中给出对话响应。

Long-range context modeling is crucial to both dialogue understanding and generation. The most popular method for dialogue context representation is to concatenate the last-$k$ previous utterances. However, this method may not be ideal for conversations containing long-range dependencies. In this work, we propose DialoGX, a novel encoder-decoder based framework for conversational response generation with a generalized and explainable context representation that can look beyond the last-$k$ utterances. Hence the method is adaptive to conversations with long-range dependencies. The main idea of our approach is to identify and utilize the most relevant historical utterances instead of the last-$k$ utterances in chronological order. We study the effectiveness of our proposed method on both dialogue generation (open-domain) and understanding (DST) tasks. DialoGX achieves comparable performance with the state-of-the-art models on DailyDialog dataset. We also observe performance gain in existing DST models with our proposed context representation strategy on MultiWOZ dataset. We justify our context representation through the lens of psycholinguistics and show that the relevance score of previous utterances agrees well with human cognition which makes DialoGX explainable as well.

Standard language model training employs gold human documents or human-human interaction data, and treats all training data as positive examples. Growing evidence shows that even with very large amounts of positive training data, issues remain that can be alleviated with relatively small amounts of negative data -- examples of what the model should not do. In this work, we propose a novel procedure to train with such data called the CRINGE loss (ContRastive Iterative Negative GEneration). We show the effectiveness of this approach across three different experiments on the tasks of safe generation, contradiction avoidance, and open-domain dialogue. Our models outperform multiple strong baselines and are conceptually simple, easy to train and implement.

我们介绍了Sparrow，这是一个寻求信息的对话代理，与提示的语言模型基线相比，训练有素，更有帮助，正确和无害。我们使用从人类反馈中的强化学习来培训我们的模型，以帮助人类评估者判断代理人的行为。首先，为了使我们的代理人更有帮助和无害，我们将良好对话的要求分解为代理人应遵循的自然语言规则，并分别向评估者询问每个规则。我们证明，这种崩溃使我们能够收集对代理行为的更多针对性的人类判断，并允许更有效的规则条件奖励模型。其次，我们的代理商在收集对模型声明的偏好判决时提供了支持事实主张的来源的证据。对于事实问题，麻雀提供的证据支持了78％的时间。比基线比基线更享受麻雀，同时对人类的对抗性探测更具弹性，在探测时只有8％的时间违反了我们的规则。最后，我们进行了广泛的分析，表明尽管我们的模型学会遵守我们的规则，但它可以表现出分布偏见。

尽管改善神经对话代理的事实准确性是大量研究的对象，但在神经对话的环境中，沟通的另一个重要方面是对无知的透明度。在这项工作中，我们分析了最新的聊天模型在多大程度上是语言校准的，因为它们的疑问（或信心）的口头表达与该模型的响应实际上是不正确（或正确）的可能性相匹配。 。我们发现这些模型的校准很差，但是我们表明可以准确预测正确性的可能性。通过将这种元认知特征纳入可控生成模型的训练中，我们获得了具有大大改进语言校准的对话代理。尽管改善神经对话代理的事实准确性是大量研究的对象，但在神经对话的环境中，沟通的另一个重要方面是对无知的透明度。在这项工作中，我们分析了最新的聊天模型在多大程度上是语言校准的，因为它们的疑问（或信心）的口头表达与该模型的响应实际上是不正确（或正确）的可能性相匹配。 。我们发现这些模型的校准很差，但是我们表明可以准确预测正确性的可能性。通过将这种元认知特征纳入可控生成模型的训练中，我们获得了具有大大改进语言校准的对话代理。

预培训语言模型的浪潮一直不断提高机器生成的对话的质量，然而，一些产生的响应仍然遭受过度重复，有时重复从话语中重复单词，有时重复自我产生的响应中的单词，或者两个都。不当重复单词可以显着降低生成文本的质量。受到惩罚的采样是一种流行的解决方案，减少了推理期间现有词的采样概率，但是，它非常容易受到静态的不适当的设置。将其设置得太高可以产生奇怪和不切实际的句子，同时将其设置得太低，使得抑制重复微不足道的任务。要解决上述方法的缺点，我们设计了一个上下文感知的分类器，以明确决定何时允许重复和何时采用惩罚的采样。这种分类器可以容易地与现有的解码方法集成，在保持文本的分集的同时在适当的情况下减少重复。实验结果表明，我们的方法可以产生更高质量和更真实的对话。

许多古典童话，小说和剧本都利用对话来推进故事情节并建立角色。我们提出了第一个研究，以探索机器是否可以理解和产生故事中的对话，这需要捕获不同角色的特征及其之间的关系。为此，我们提出了两项​​新任务，包括蒙版对话生成和对话演讲者的认可，即分别产生对话转弯和预测说话者的指定对话转弯。我们构建了一个新的数据集拨号故事，该数据集由105K中国故事组成，其中包含大量对话，以支持评估。我们通过对拨号故事进行自动和手动评估测试现有模型来显示提出的任务的困难。此外，我们建议学习明确的角色表示，以提高这些任务的绩效。广泛的实验和案例研究表明，我们的方法可以产生更连贯和信息丰富的对话，并获得比强基础更高的说话者识别精度。

个性化对话代理对于对话系统非常重要，以产生更具体，一致，并从事和吸引力的反应。然而，大多数当前对话的个性化方法依赖于推理期间的明确人物描述，严重限制其应用。在本文中，我们提出了一种新颖的方法，该方法将根据对话历史来预测人物信息，以个性化对话代理而不依赖于推理期间的任何明确的人格描述。 Personachat数据集上的实验结果表明，当在对话剂的预测轮廓上调节（即“自身角色”）时，所提出的方法可以提高所产生的响应的一致性，并在预测的角色调节时改善所产生的响应的接合对话伙伴（即“他们的角色”）。我们还发现培训的角色预测模型可以成功转移到其他数据集，并帮助生成更相关的响应。

我们介绍了AARGH，这是一个面向任务的对话框系统，该系统结合了单个模型中的检索和生成方法，旨在改善对话框管理和输出的词汇多样性。该模型采用了一种新的响应选择方法，该方法基于动作感知训练目标和简化的单编码检索架构，该方法使我们能够构建端到端检索增强生成模型，在该模型中，检索和生成共享大多数参数。在Multiwoz数据集上，我们表明我们的方法与最先进的基线相比，在维持或改善状态跟踪和上下文响应生成性能的同时，产生了更多的输出。

Due to the lack of human resources for mental health support, there is an increasing demand for employing conversational agents for support. Recent work has demonstrated the effectiveness of dialogue models in providing emotional support. As previous studies have demonstrated that seekers' persona is an important factor for effective support, we investigate whether there are benefits to modeling such information in dialogue models for support. In this paper, our empirical analysis verifies that persona has an important impact on emotional support. Therefore, we propose a framework for dynamically inferring and modeling seekers' persona. We first train a model for inferring the seeker's persona from the conversation history. Accordingly, we propose PAL, a model that leverages persona information and, in conjunction with our strategy-based controllable generation method, provides personalized emotional support. Automatic and manual evaluations demonstrate that our proposed model, PAL, achieves state-of-the-art results, outperforming the baselines on the studied benchmark. Our code and data are publicly available at https://github.com/chengjl19/PAL.

We have a Christmas gift for Harry Potter fans all over the world. In this paper, we present Harry Potter Dialogue (HPD), a dataset that helps train Harry Potter-like dialogue agents. Such a task is typically viewed as a variant of personalized dialogue agents, but they differ significantly in three respects: 1) Harry lived in a virtual world of wizards, thus, real-world commonsense may not apply to Harry's conversations; 2) Harry's behavior is strongly linked to background information in conversations: the scene, its attributes and its relationship to other speakers; and 3) Such backgrounds are dynamically altered as the storyline goes on. The HPD dataset, as the first dataset to facilitate the study of dialogue agent construction for characters within a story, provides rich contextual information about each dialogue session such as scenes, character attributes, and relations. More importantly, all the background information will change over the course of the story. In addition, HPD could support both dialogue generation and retrieval tasks. We evaluate baselines such as Dialog-GPT and BOB to determine the extent to which they can generate Harry Potter-like responses. The experimental results disappoint us in that although the generated responses are fluent, they still seem out of character for Harry. Besides, we validate the current most robust dialogue agent, ChatGPT, which also can't generate plausible Harry-Potter-like responses in some cases, either. Our results suggest that there is much scope for future research.