We study knowledge-grounded dialogue generation with pre-trained language models. Instead of pursuing new state-of-the-art on benchmarks, we try to understand if the knowledge stored in parameters of the pre-trained models is already enough to ground open domain dialogues, and thus allows us to get rid of the dependency on external knowledge sources in generation. Through extensive experiments on benchmarks, we find that by fine-tuning with a few dialogues containing knowledge, the pre-trained language models can outperform the state-of-the-art model that requires external knowledge in automatic evaluation and human judgment, suggesting a positive answer to the question we raised.
Multi-choice machine reading comprehension (MRC) requires models to choose the correct answer from candidate options given a passage and a question. Our research focuses dialogue-based MRC, where the passages are multi-turn dialogues. It suffers from two challenges, the answer selection decision is made without support of latently helpful commonsense, and the multi-turn context may hide considerable irrelevant information. This work thus makes the first attempt to tackle those two challenges by extracting substantially important turns and utilizing external knowledge to enhance the representat…
Existing knowledge-grounded dialogue systems typically use finetuned versions of a pretrained language model (LM) and large-scale knowledge bases. These models typically fail to generalize on topics outside of the knowledge base, and require maintaining separate potentially large checkpoints each time finetuning is needed. In this paper, we aim to address these limitations by leveraging the inherent knowledge stored in the pretrained LM as well as its powerful generation ability. We propose a multi-stage prompting approach to generate knowledgeable responses from a single pretrained LM. We fir…
University of Potsdam（波茨坦大学）；Volkswagen AG（大众汽车公司）
Fully data driven Chatbots for non-goal oriented dialogues are known to suffer from inconsistent behaviour across their turns, stemming from a general difficulty in controlling parameters like their assumed background personality and knowledge of facts. One reason for this is the relative lack of labeled data from which personality consistency and fact usage could be learned together with dialogue behaviour. To address this, we introduce a new labeled dialogue dataset in the domain of movie discussions, where every dialogue is based on pre-specified facts and opinions. We thoroughly validate t…
ChatGPT is a Knowledgeable but Inexperienced Solver: An Investigation of Commonsense Problem in Large Language Models
University of Chinese Academy of Sciences（中国科学院大学）；Chinese Information Processing Laboratory（中文信息处理实验室）；Ricoh Software Research Center (Beijing) Co., Ltd.（理光软件研究中心（北京）有限公司）
Are LLMs Really Not Knowledgeable? Mining the Submerged Knowledge in LLMs' Memory
The Hong Kong University of Science and Technology (Guangzhou)（香港科学与技术大学（广州））；The Hong Kong University of Science and Technology（香港科学与技术大学）；University of California, Merced（加州大学默塞德分校）；The University of Queensland（昆士兰大学）
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 rele…
Knowledge-driven dialog system has recently made remarkable breakthroughs. Compared with general dialog systems, superior knowledge-driven dialog systems can generate more informative and knowledgeable responses with pre-provided knowledge. However, in practical applications, the dialog system cannot be provided with corresponding knowledge in advance because it cannot know in advance the development of the conversation. Therefore, in order to make the knowledge dialogue system more practical, it is vital to find a way to retrieve relevant knowledge based on the dialogue history. To solve this…
Knowledgeable In-Context Tuning: Exploring and Exploiting Factual Knowledge for In-Context Learning
School of Data Science and Engineering, East China Normal University（华东师范大学数据科学与工程学院）；Alibaba Group（阿里巴巴集团）；School of Statistics, East China Normal University（华东师范大学统计学院）
Facebook AI Research（Facebook人工智能研究院）；Université de Lorraine（洛林大学）；LORIA（洛林计算机科学及其应用研究实验室）；CNRS（法国国家科学研究中心）；IRIT（图卢兹计算机研究所）
Various machine learning tasks can benefit from access to external information of different modalities, such as text and images. Recent work has focused on learning architectures with large memories capable of storing this knowledge. We propose augmenting generative Transformer neural networks with KNN-based Information Fetching (KIF) modules. Each KIF module learns a read operation to access fixed external knowledge. We apply these modules to generative dialog modeling, a challenging task where information must be flexibly retrieved and incorporated to maintain the topic and flow of conversat…
Abstractive dialogue summarization has long been viewed as an important standalone task in natural language processing, but no previous work has explored the possibility of whether abstractive dialogue summarization can also be used as a means to boost an NLP system's performance on other important dialogue comprehension tasks. In this paper, we propose a novel type of dialogue summarization task - STRUctured DiaLoguE Summarization - that can help pre-trained language models to better understand dialogues and improve their performance on important dialogue comprehension tasks. We further colle…
IBM Research China（IBM中国研究院）；Beijing Institute of Technology（北京理工大学）
In end-to-end dialogue modeling and agent learning, it is important to (1) effectively learn knowledge from data, and (2) fully utilize heterogeneous information, e.g., dialogue act flow and utterances. However, the majority of existing methods cannot simultaneously satisfy the two conditions. For example, rule definition and data labeling during system design take too much manual work, and sequence-to-sequence methods only model one-side utterance information. In this paper, we propose a novel joint end-to-end model by multi-task representation learning, which can capture the knowledge from h…
Synthesize-on-Graph: Knowledgeable Synthetic Data Generation for Continue Pre-training of Large Language Models
DataArc Tech Ltd.（DataArc科技有限公司）；IDEA Research, International Digital Economy Academy（IDEA研究所、国际数字经济学院）；Gaoling School of Artificial Intelligence, Renmin University of China（法律人工智能学院，中国人民大学）
Shanghai Jiao Tong University（上海交通大学）；Microsoft Research Asia（微软亚洲研究院）；AISpeech Co（思必驰公司）
Building a universal conversational agent has been a long-standing goal of the dialogue research community. Most previous works only focus on a small set of dialogue tasks. In this work, we aim to build a unified dialogue foundation model (DFM) which can be used to solve massive diverse dialogue tasks. To achieve this goal, a large-scale well-annotated dialogue dataset with rich task diversity (DialogZoo) is collected. We introduce a framework to unify all dialogue tasks and propose novel auxiliary self-supervised tasks to achieve stable training of DFM on the highly diverse large scale Dialog…
Shanghai Jiao Tong University（上海交通大学）；China Merchants Bank Credit Card Center（招商银行信用卡中心）
Previous dialogue summarization techniques adapt large language models pretrained on the narrative text by injecting dialogue-specific features into the models. These features either require additional knowledge to recognize or make the resulting models harder to tune. To bridge the format gap between dialogues and narrative summaries in dialogue summarization tasks, we propose to post-train pretrained language models (PLMs) to rephrase from dialogue to narratives. After that, the model is fine-tuned for dialogue summarization as usual. Comprehensive experiments show that our approach signific…
Multi-turn dialogue modeling as a challenging branch of natural language understanding (NLU), aims to build representations for machines to understand human dialogues, which provides a solid foundation for multiple downstream tasks. Recent studies of dialogue modeling commonly employ pre-trained language models (PrLMs) to encode the dialogue history as successive tokens, which is insufficient in capturing the temporal characteristics of dialogues. Therefore, we propose Bidirectional Information Decoupling Network (BiDeN) as a universal dialogue encoder, which explicitly incorporates both the p…
Pre-trained language models (PrLM) has been shown powerful in enhancing a broad range of downstream tasks including various dialogue related ones. However, PrLMs are usually trained on general plain text with common language model (LM) training objectives, which cannot sufficiently capture dialogue exclusive features due to the limitation of such training setting, so that there is an immediate need to fill the gap between a specific dialogue task and the LM task. As it is unlikely to collect huge dialogue data for dialogue-oriented pre-training, in this paper, we propose three strategies to si…
National University of Singapore（新加坡国立大学）；Universidad Politécnica de Madrid（马德里理工大学）；Robert Bosch (SEA) Pte Ltd（博世（东南亚）私人有限公司）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
Chatbots are expected to be knowledgeable across multiple domains, e.g. for daily chit-chat, exchange of information, and grounding in emotional situations. To effectively measure the quality of such conversational agents, a model-based automatic dialogue evaluation metric (ADEM) is expected to perform well across multiple domains. Despite significant progress, an ADEM that works well in one domain does not necessarily generalize to another. This calls for a dedicated network architecture for domain generalization. To tackle the multi-domain dialogue evaluation task, we propose a Panel of Expe…
Renmin University of China（中国人民大学）；Peking University（北京大学）；Microsoft Corporation（微软公司）
Knowledge-grounded conversation (KGC) shows great potential in building an engaging and knowledgeable chatbot, and knowledge selection is a key ingredient in it. However, previous methods for knowledge selection only concentrate on the relevance between knowledge and dialogue context, ignoring the fact that age, hobby, education and life experience of an interlocutor have a major effect on his or her personal preference over external knowledge. Without taking the personalization issue into account, it is difficult to select the proper knowledge and generate persona-consistent responses. In thi…
AI Pedagogy: Dialogic Social Learning for Artificial Agents
University of Milan-Bicocca（米兰-比科卡大学）；Technical University of Munich（慕尼黑技术大学）；Institute for Cognitive Sciences and Technologies (ISTC-CNR)（认知科学与技术研究所（ISTC-CNR））
受维果茨基社会文化理论启发，本研究提出AI Social Gym动态环境，通过AI师生智能体间的二元教学对话进行本体获取，证明混合方向交互显著提升LLM知识获取与应用能力，为AI训练提供补充路径。
ETH Zürich（苏黎世联邦理工学院）；Facebook AI Research（Facebook人工智能研究院）
Large language models can produce fluent dialogue but often hallucinate factual inaccuracies. While retrieval-augmented models help alleviate this issue, they still face a difficult challenge of both reasoning to provide correct knowledge and generating conversation simultaneously. In this work, we propose a modular model, Knowledge to Response (K2R), for incorporating knowledge into conversational agents, which breaks down this problem into two easier steps. K2R first generates a knowledge sequence, given a dialogue context, as an intermediate step. After this "reasoning step", the model then…
Embedding-Informed Adaptive Retrieval-Augmented Generation of Large Language Models
University of New South Wales（新南威尔士大学）；University of California San Diego（加利福尼亚大学圣迭戈分校）；Duke University（杜克大学）；Georgia Institute of Technology（佐治亚理工学院）；Adobe Research（奥多比研究院）；CSIRO’s Data61（联邦科学与工业研究组织Data61）
"In Dialogues We Learn": Towards Personalized Dialogue Without Pre-defined Profiles through In-Dialogue Learning
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；Ant Group（蚂蚁集团）；University of Electronic Science and Technology of China（电子科技大学）；Kunming University of Science and Technology（昆明理工大学）
With the rapid development of artificial intelligence, conversational bots have became prevalent in mainstream E-commerce platforms, which can provide convenient customer service timely. To satisfy the user, the conversational bots need to understand the user's intention, detect the user's emotion, and extract the key entities from the conversational utterances. However, understanding dialogues is regarded as a very challenging task. Different from common language understanding, utterances in dialogues appear alternately from different roles and are usually organized as hierarchical structures…
Pre-trained language models (PrLMs) have demonstrated superior performance due to their strong ability to learn universal language representations from self-supervised pre-training. However, even with the help of the powerful PrLMs, it is still challenging to effectively capture task-related knowledge from dialogue texts which are enriched by correlations among speaker-aware utterances. In this work, we present SPIDER, Structural Pre-traIned DialoguE Reader, to capture dialogue exclusive features. To simulate the dialogue-like features, we propose two training objectives in addition to the ori…
Shanghai Jiao Tong University（上海交通大学）；MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University（教育部人工智能重点实验室、上海交通大学人工智能研究院）
Training machines to understand natural language and interact with humans is an elusive and essential task of artificial intelligence. A diversity of dialogue systems has been designed with the rapid development of deep learning techniques, especially the recent pre-trained language models (PrLMs). Among these studies, the fundamental yet challenging type of task is dialogue comprehension whose role is to teach the machines to read and comprehend the dialogue context before responding. In this paper, we review the previous methods from the technical perspective of dialogue modeling for the dia…
Training machines to understand natural language and interact with humans is an elusive and essential task of artificial intelligence. A diversity of dialogue systems has been designed with the rapid development of deep learning techniques, especially the recent pre-trained language models (PrLMs). Among these studies, the fundamental yet challenging type of task is dialogue comprehension whose role is to teach the machines to read and comprehend the dialogue context before responding. In this paper, we review the previous methods from the technical perspective of dialogue modeling for the dia…
We present MixingBoard, a platform for quickly building demos with a focus on knowledge grounded stylized text generation. We unify existing text generation algorithms in a shared codebase and further adapt earlier algorithms for constrained generation. To borrow advantages from different models, we implement strategies for cross-model integration, from the token probability level to the latent space level. An interface to external knowledge is provided via a module that retrieves on-the-fly relevant knowledge from passages on the web or any document collection. A user interface for local deve…
Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）；Meituan（美团）；International Innovation Center of Tsinghua University（清华大学国际创新中心）
Tuning pre-trained language models (PLMs) with task-specific prompts has been a promising approach for text classification. Particularly, previous studies suggest that prompt-tuning has remarkable superiority in the low-data scenario over the generic fine-tuning methods with extra classifiers. The core idea of prompt-tuning is to insert text pieces, i.e., template, to the input and transform a classification problem into a masked language modeling problem, where a crucial step is to construct a projection, i.e., verbalizer, between a label space and a label word space. A verbalizer is usually …
Chinese Academy of Sciences（中国科学院）；Institute of Computing Technology, CAS（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Tencent（腾讯）
In this study, we focus on extracting knowledgeable snippets and annotating knowledgeable documents from Web corpus, consisting of the documents from social media and We-media. Informally, knowledgeable snippets refer to the text describing concepts, properties of entities, or relations among entities, while knowledgeable documents are the ones with enough knowledgeable snippets. These knowledgeable snippets and documents could be helpful in multiple applications, such as knowledge base construction and knowledge-oriented service. Previous studies extracted the knowledgeable snippets using the…
Tsinghua University（清华大学）；Renmin University of China（中国人民大学）；BNRist（北京国家信息科学与技术研究中心）；Institute for Artificial Intelligence, Tsinghua University（清华大学人工智能研究院）
Knowledge graphs, as the cornerstone of many AI applications, usually face serious incompleteness problems. In recent years, there have been many efforts to study automatic knowledge graph completion (KGC), most of which use existing knowledge to infer new knowledge. However, in our experiments, we find that not all relations can be obtained by inference, which constrains the performance of existing models. To alleviate this problem, we propose a new model based on information retrieval and reading comprehension, namely IR4KGC. Specifically, we pre-train a knowledge-based information retrieval…
Dialogue systems, also called chatbots, are now used in a wide range of applications. However, they still have some major weaknesses. One key weakness is that they are typically trained from manually-labeled data and/or written with handcrafted rules, and their knowledge bases (KBs) are also compiled by human experts. Due to the huge amount of manual effort involved, they are difficult to scale and also tend to produce many errors ought to their limited ability to understand natural language and the limited knowledge in their KBs. Thus, the level of user satisfactory is often low. In this pape…
Wangxuan Institute of Computer Technology, Peking University（北京大学王选计算机研究所）；Center for Data Science, AAIS, Peking University（北京大学人工智能研究院数据科学中心）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；Microsoft Corporation（微软公司）；Meituan（美团）
Grounding dialogue generation by extra knowledge has shown great potentials towards building a system capable of replying with knowledgeable and engaging responses. Existing studies focus on how to synthesize a response with proper knowledge, yet neglect that the same knowledge could be expressed differently by speakers even under the same context. In this work, we mainly consider two aspects of knowledge expression, namely the structure of the response and style of the content in each part. We therefore introduce two sequential latent variables to represent the structure and the content style…
University of Southern California（南加州大学）；Information Sciences Institute, University of Southern California（南加州大学信息科学研究所）；University of California, Los Angeles（加州大学洛杉矶分校）
Commonsense question answering (QA) requires background knowledge which is not explicitly stated in a given context. Prior works use commonsense knowledge graphs (KGs) to obtain this knowledge for reasoning. However, relying entirely on these KGs may not suffice, considering their limited coverage and the contextual dependence of their knowledge. In this paper, we augment a general commonsense QA framework with a knowledgeable path generator. By extrapolating over existing paths in a KG with a state-of-the-art language model, our generator learns to connect a pair of entities in text with a dy…
Have you ever looked at a painting and wondered what is the story behind it? This work presents a framework to bring art closer to people by generating comprehensive descriptions of fine-art paintings. Generating informative descriptions for artworks, however, is extremely challenging, as it requires to 1) describe multiple aspects of the image such as its style, content, or composition, and 2) provide background and contextual knowledge about the artist, their influences, or the historical period. To address these challenges, we introduce a multi-topic and knowledgeable art description framew…
University of Chinese Academy of Sciences（中国科学院大学）；Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所）；Tencent Inc.（腾讯公司）
Previous literatures show that pre-trained masked language models (MLMs) such as BERT can achieve competitive factual knowledge extraction performance on some datasets, indicating that MLMs can potentially be a reliable knowledge source. In this paper, we conduct a rigorous study to explore the underlying predicting mechanisms of MLMs over different extraction paradigms. By investigating the behaviors of MLMs, we find that previous decent performance mainly owes to the biased prompts which overfit dataset artifacts. Furthermore, incorporating illustrative cases and external contexts improve kn…
NetEase Inc.（网易公司）；Tsinghua University（清华大学）；Samsung Research China - Beijing（三星电子中国研究院（北京））；Alibaba Group（阿里巴巴集团）
Recent advances in open-domain dialogue systems rely on the success of neural models that are trained on large-scale data. However, collecting large-scale dialogue data is usually time-consuming and labor-intensive. To address this data dilemma, we propose a novel data augmentation method for training open-domain dialogue models by utilizing unpaired data. Specifically, a data-level distillation process is first proposed to construct augmented dialogues where both post and response are retrieved from the unpaired data. A ranking module is employed to filter out low-quality dialogues. Further, …
Nanjing University of Aeronautics and Astronautics（南京航空航天大学）；Shandong University（山东大学）
Dialogue structure discovery is essential in dialogue generation. Well-structured topic flow can leverage background information and predict future topics to help generate controllable and explainable responses. However, most previous work focused on dialogue structure learning in task-oriented dialogue other than open-domain dialogue which is more complicated and challenging. In this paper, we present a new framework CTRLStruct for dialogue structure learning to effectively explore topic-level dialogue clusters as well as their transitions with unlabelled information. Precisely, dialogue utte…
