Dialogue Systems Research Articles

MeDi‐TODER: Medical Domain‐Incremental Task‐Oriented Dialogue Generator Using Experience Replay

ABSTRACTArtificial intelligence (AI) technology has brought groundbreaking changes to the healthcare domain. Specifically, the integration of a medical dialogue system (MDS) has facilitated interactions with patients, identifying meaningful information such as symptoms and medications from their dialogue history to generate appropriate responses. However, shortcomings arise when MDS lacks access to the patient's cumulative history or prior domain knowledge, resulting in the generation of inaccurate responses. To address this challenge, we propose a medical domain‐incremental task‐oriented dialogue generator using experience replay (MeDi‐TODER) that applies the continual learning technique to the medical task‐oriented dialogue generator. By strategically sampling and storing exemplars from previous domains and rehearsing it as it learns, the model effectively retains knowledge and can respond to the novel domains. Extensive experiments demonstrated that MeDi‐TODER significantly outperforms other models that lack continual learning in both natural language generation and natural language understanding. Notably, our proposed method achieves the highest scores, surpassing the upper‐class benchmarks.

From Function to Relation: Exploring the Dual Influences of Warmth and Competence on Generative Artificial Intelligence Services in the Hospitality Industry

Generative artificial intelligence (GenAI) has emerged as a transformative force in the rapidly-evolving hospitality industry. This study examined the correlation of acceptance of GenAI services with perceived warmth and competence, and the consequent correlations with customer outcomes in the hospitality industry. Employing a Hospitality Service Dialogue System prototype, this study collected responses from 1,306 hotel guests. Perceptions of warmth and competence correlated with increased customer intention to use GenAI services and satisfaction. Satisfaction correlated significantly with customer–brand identification and attachment. The Technology Readiness Index moderated the relationships of warmth and competence with the acceptance of GenAI services. This investigation extends the stereotype content model and the behaviors from intergroup affect and stereotypes map framework by integrating them with the technology acceptance literature, unveiling service innovations enabled by technology in the hospitality industry. This study provides hotel managers, technology developers, and marketers with strategic directions for implementing GenAI.

The Art of Dialogue: Synergistic Effects of Language Adaptation and Emotional Intelligence in AI Systems

Artificial intelligence (AI) systems are increasingly applied in areas such as voice assistants, customer service, and healthcare. However, current AI systems often lack semantic understanding and emotional expression capabilities when interacting with humans, leading to poor user experiences. To enable AI systems to converse more naturally with humans, it is necessary to synergistically optimize two key capabilities: language adaptation and emotional intelligence. This paper reviews the latest research progress in AI dialogue systems regarding language adaptation and emotional intelligence. In terms of language adaptation, this paper summarizes technologies such as personalized dialogue, knowledge-enhanced dialogue, and multi-turn contextual dialogue. In terms of emotional intelligence, this paper outlines work in areas such as emotion recognition, emotional dialogue generation, and empathetic dialogue. Additionally, this paper discusses strategies for synergistic optimization of language adaptation and emotional intelligence, as well as challenges in dialogue system evaluation. Finally, this paper looks ahead to future directions for AI dialogue systems.

Deep Learning Applications in Natural Language Processing and Optimization Strategies

In recent years, deep learning has made significant advancements in the field of natural language processing (NLP), driving the development of language understanding and generation tasks. Deep learning utilizes multi-layer neural networks to automatically extract and represent complex data features. This paper first introduces the basic principles of deep learning and mainstream frameworks (such as TensorFlow and PyTorch) and discusses core NLP tasks, including word embedding, language modeling, and text generation. Subsequently, it analyzes specific practices of deep learning applications in text classification, machine translation, automatic summarization, dialogue systems, and speech recognition. Further discussion covers model optimization methods, including structural optimization (such as RNN, LSTM, GRU, and Transformer), data preprocessing and feature engineering, hyperparameter tuning, and accelerated computation. Finally, it explores cutting-edge optimization strategies like federated learning, model compression, self-supervised learning, and transfer learning, proposing future research directions and challenges. This paper aims to systematically analyze the applications and optimization strategies of deep learning in NLP.

ChatGPT as Teacher Assistant for Physics Teaching

This study explores the integration of ChatGPT as a teaching assistant in physics education, emphasizing its potential to transform traditional pedagogical approaches. ChatGPT facilitates interactive and inquiry-based learning grounded in constructivist learning theory, allowing students to engage actively in experiments and better grasp abstract concepts through hands-on activities. The AI's adaptive dialogue systems promote socio-constructivist learning by encouraging social interaction and personalized feedback, which is essential for addressing individual learning gaps and enhancing student engagement. ChatGPT's ability to simulate real-world physics problems and provide immediate feedback fosters experiential learning, making complex concepts more accessible and promoting critical thinking skills. By offering tailored interventions and adapting to individual learning paces, ChatGPT supports a personalized educational experience that caters to the unique needs of each student. This adaptability is particularly beneficial in physics education, where students often struggle with abstract concepts and require immediate clarification to progress effectively. The paper concludes that the synthesis of generative AI and pedagogy has the potential to reshape science education, fostering deeper understanding and curiosity among students. By leveraging innovative methodologies and augmented data, ChatGPT enriches teacher-student interactions, creating a comprehensive educational experience that promotes a culture of curiosity and exploration, thereby nurturing future scientists and engineers. Ultimately, the integration of ChatGPT into physics education offers a valuable opportunity to enhance student engagement and understanding of scientific concepts through interactive and personalized support.

DialTest‐EA: An Enhanced Fuzzing Approach With Energy Adjustment for Dialogue Systems via Metamorphic Testing

DialTest‐EA: An Enhanced Fuzzing Approach With Energy Adjustment for Dialogue Systems via Metamorphic Testing

The role of ChatGPT on academic research: perspectives from filipino students across diverse educational levels

With the emergence of technology, artificial intelligence (AI) is introducing a new wave of remarkable tools and transformations in education. One such technological development is ChatGPT, a dialogue system designed to collect and organize information, which is then converted into text-based inputs to address user inquiries. There is also a clear trend in which many Filipino students increasingly use ChatGPT for academic research. Filipino students' growing usage of ChatGPT demands immediate attention to examine its role in academic research within the Philippine educational setting. Thus, this study examines how ChatGPT is used in academic research by exploring students' perspectives and analyzing its perceived benefits and challenges. A descriptive cross-sectional design was conducted in the Eastern Visayas region of the Philippines, explicitly targeting selected secondary and tertiary schools. A total of 316 students were involved in the study, selected through convenience sampling. Results revealed ChatGPT to be moderately useful for various aspects of academic research and moderately influential in the student's academic performance. Receiving additional information or resources, getting help with mechanics, writing, paraphrasing, and generating ideas were found to be the benefits of using ChatGPT. Nonetheless, students commonly encountered issues such as redundant, inadequate, and inaccurate information provided by ChatGPT and concerns regarding its accessibility and the potential for plagiarism. To optimize advantages and address possible concerns, academic institutions should establish concise guidelines and regulations for responsible student utilization of ChatGPT

Joint intent detection and slot filling with syntactic and semantic features using multichannel CNN-BiLSTM

Understanding spoken language is crucial for conversational agents, with intent detection and slot filling being the primary tasks in natural language understanding (NLU). Enhancing the NLU tasks can lead to an accurate and efficient virtual assistant thereby reducing the need for human intervention and expanding their applicability in other domains. Traditionally, these tasks have been addressed individually, but recent studies have highlighted their interconnection, suggesting better results when solved together. Recent advances in natural language processing have shown that pretrained word embeddings can enhance text representation and improve the generalization capabilities of models. However, the challenge of poor generalization in joint learning models for intent detection and slot filling remains due to limited annotated datasets. Additionally, traditional models face difficulties in capturing both the semantic and syntactic nuances of language, which are vital for accurate intent detection and slot filling. This study proposes a hybridized text representation method using a multichannel convolutional neural network with three embedding channels: non-contextual embeddings for semantic information, part-of-speech (POS) tag embeddings for syntactic features, and contextual embeddings for deeper contextual understanding. Specifically, we utilized word2vec for non-contextual embeddings, one-hot vectors for POS tags, and bidirectional encoder representations from transformers (BERT) for contextual embeddings. These embeddings are processed through a convolutional layer and a shared bidirectional long short-term memory (BiLSTM) network, followed by two softmax functions for intent detection and slot filling. Experiments on the air travel information system (ATIS) and SNIPS datasets demonstrated that our model significantly outperformed the baseline models, achieving an intent accuracy of 97.90% and slot filling F1-score of 98.86% on the ATIS dataset, and an intent accuracy of 98.88% and slot filling F1-score of 97.07% on the SNIPS dataset. These results highlight the effectiveness of our proposed approach in advancing dialogue systems, and paving the way for more accurate and efficient natural language understanding in real-world applications.

Economic aspects of error identification in semi-structured publications in the state language

Due to the rapid growth of information on the Internet and social networks, research in the field of computational linguistics has become very relevant. The volume of information that people and machines create in natural language needs to be processed, analyzed and verified. Information retrieval systems, dialog systems, and machine translation tools are used for this. The range of automatic text processing systems is very wide, it covers various tasks. Finding errors in texts and words, identifying and correcting incorrect words is one of the most important tasks of natural language processing (NLP). The article provides an overview of semi-structured data, methods and technologies for identifying incorrect words in natural languages. The paper gives an overview of semi-structured data, methods and techniques for detecting incorrect words in natural languages. The aim of the research is to develop an effective approach for detecting and correcting errors occurring in Kazakh-language texts, especially in the context of limited resources and unstructured data. The research includes the use of machine learning techniques as well as economic analysis of the costs of developing and implementing such solutions. The proposed approach facilitates the automation of text verification, which can significantly reduce the cost of manual data processing and improve the quality of information in various spheres, including business and public administration.

CIRG-SL: Commonsense Inductive Relation Graph framework with Soft Labels for Empathetic Response Generation

Empathy, the ability to understand and respond to others’ emotions, is critical in dialogue systems, particularly for applications such as psychological counselling and casual conversation. However, existing approaches often struggle to accurately capture emotional transitions in user utterances and dialogue contexts across multiple levels of granularity. Additionally, the orthogonality of one-hot emotion representations overlooks the subtle relationships between emotion categories. To overcome these limitations, we propose a novel framework: the commonsense inductive relation graph with fine emotional soft labels for empathetic response generation. This framework refines the representation of user utterances by incorporating commonsense knowledge to derive external emotional cues, which are integrated into an adaptive inductive relation graph. This graph explicitly captures and models dynamic emotional trajectories throughout the dialogue. Additionally, we propose a fine-grained soft label construction method that more accurately reflects the nuanced distribution of emotions, replacing the rigid one-hot encoding with a continuous representation that better captures inter-emotion relationships. Our approach achieves strong performance on the benchmark EmpatheticDialogues dataset, with a perplexity score of 27.85, Dist-1 of 1.81, Dist-2 of 6.58, and a dialogue emotion accuracy of 55.65%. These results highlight the effectiveness of our framework in capturing subtle emotions and generating contextually rich, empathetic responses.

An Emotional Dialogue System Using Conditional Generative Adversarial Networks with a Sequence-to-Sequence Transformer Encoder

Understanding the expression of emotion and generating appropriate responses are key steps toward constructing emotional, conversational agents. In this paper, we propose a framework for single-turn emotional conversation generation, and there are three main components in our model, namely, a sequence-to-sequence model with stacked encoders, a conditional variational autoencoder, and conditional generative adversarial networks. For the sequence-to-sequence model with stacked encoders, we designed a two-layer encoder by combining Transformer with gated recurrent units-based neural networks. Because of the flexibility of the sequence-to-sequence model, we adopted a conditional variational autoencoder in our framework, which uses latent variables to learn a distribution over potential responses and generates diverse responses. Furthermore, we regard a conditional variational autoencoder-based, sequence-to-sequence model as the generative model, and the training of the generative model is assisted by both a content discriminator and an emotion classifier, which assists our model in promoting content information and emotion expression. We use automated evaluation and human evaluation to evaluate our model and baselines on the NTCIR short text conversation task (STC-3) Chinese emotional conversation generation (CECG) Subtask dataset [44], and the experimental results demonstrate that our proposed framework can generate semantically reasonable and emotionally appropriate responses.

Combining replay and LoRA for continual learning in natural language understanding

Large language models have significantly improved dialogue systems through enhanced capabilities in understanding queries and generating responses. Despite these enhancements, task-oriented dialogue systems—which power many intelligent assistants—face challenges when adapting to new domains and applications. This challenge arises from a phenomenon known as catastrophic forgetting, where models forget previously acquired knowledge when learning new tasks. This paper addresses this issue through continual learning techniques to preserve previously learned knowledge while seamlessly integrating new tasks and domains. We propose Experience Replay Informative-Low Rank Adaptation or ERI-LoRA, a hybrid continual learning method for natural language understanding in dialogue systems that effectively combines the replay-based methods with parameter-efficient techniques. Our experiments on intent detection and slot-filling tasks demonstrate that ERI-LoRA significantly outperforms competitive baselines in continual learning. The results of our catastrophic forgetting experiments demonstrate that ERI-LoRA maintains robust memory stability in the model, demonstrating its effectiveness in mitigating these effects.

What’s so complex about conversational speech? A comparison of HMM-based and transformer-based ASR architectures

Highly performing speech recognition is important for more fluent human-machine interaction (e.g., dialogue systems). Modern ASR architectures achieve human-level recognition performance on read speech but still perform sub-par on conversational speech, which arguably is or, at least, will be instrumental for human-machine interaction. Understanding the factors behind this shortcoming of modern ASR systems may suggest directions for improving them. In this work, we compare the performances of HMM- vs. transformer-based ASR architectures on a corpus of Austrian German conversational speech. Specifically, we investigate how strongly utterance length, prosody, pronunciation, and utterance complexity as measured by perplexity affect different ASR architectures. Among other findings, we observe that single-word utterances – which are characteristic of conversational speech and constitute roughly 30% of the corpus – are recognized more accurately if their F0 contour is flat; for longer utterances, the effects of the F0 contour tend to be weaker. We further find that zero-shot systems require longer utterance lengths and are less robust to pronunciation variation, which indicates that pronunciation lexicons and fine-tuning on the respective corpus are essential ingredients for the successful recognition of conversational speech.

A review of reinforcement learning for natural language processing and applications in healthcare.

Reinforcement learning (RL) represents a pivotal avenue within natural language processing (NLP), offering a potent mechanism for acquiring optimal strategies in task completion. This literature review studies various NLP applications where RL has demonstrated efficacy, with notable applications in healthcare settings. To systematically explore the applications of RL in NLP, focusing on its effectiveness in acquiring optimal strategies, particularly in healthcare settings, and provide a comprehensive understanding of RL's potential in NLP tasks. Adhering to the PRISMA guidelines, an exhaustive literature review was conducted to identify instances where RL has exhibited success in NLP applications, encompassing dialogue systems, machine translation, question-answering, text summarization, and information extraction. Our methodological approach involves closely examining the technical aspects of RL methodologies employed in these applications, analyzing algorithms, states, rewards, actions, datasets, and encoder-decoder architectures. The review of 93 papers yields insights into RL algorithms, prevalent techniques, emergent trends, and the fusion of RL methods in NLP healthcare applications. It clarifies the strategic approaches employed, datasets utilized, and the dynamic terrain of RL-NLP systems, thereby offering a roadmap for research and development in RL and machine learning techniques in healthcare. The review also addresses ethical concerns to ensure equity, transparency, and accountability in the evolution and application of RL-based NLP technologies, particularly within sensitive domains such as healthcare. The findings underscore the promising role of RL in advancing NLP applications, particularly in healthcare, where its potential to optimize decision-making and enhance patient outcomes is significant. However, the ethical challenges and technical complexities associated with RL demand careful consideration and ongoing research to ensure responsible and effective implementation. By systematically exploring RL's applications in NLP and providing insights into technical analysis, ethical implications, and potential advancements, this review contributes to a deeper understanding of RL's role for language processing.

ECDG-DST: A dialogue state tracking model based on efficient context and domain guidance for smart dialogue systems

ECDG-DST: A dialogue state tracking model based on efficient context and domain guidance for smart dialogue systems

EIRAD: An Evidence-Based Dialogue System With Highly Interpretable Reasoning Path for Automatic Diagnosis

EIRAD: An Evidence-Based Dialogue System With Highly Interpretable Reasoning Path for Automatic Diagnosis

Optimizing pipeline task-oriented dialogue systems using post-processing networks

Many studies have proposed methods for optimizing the dialogue performance of an entire pipeline task-oriented dialogue system by jointly training modules in the system using reinforcement learning. However, these methods are limited in that they can only be applied to modules implemented using trainable neural-based methods. To solve this problem, we propose a method for optimizing the dialogue performance of a pipeline system that consists of modules implemented with arbitrary methods for dialogue. With our method, neural-based components called post-processing networks (PPNs) are installed inside such a system to post-process the output of each module. All PPNs are updated to improve the overall dialogue performance of the system by using reinforcement learning, not necessitating that each module be differentiable. Through dialogue simulations and human evaluations on two well-studied task-oriented dialogue datasets, CamRest676 and MultiWOZ, we show that our method can improve the dialogue performance of pipeline systems consisting of various modules. In addition, a comprehensive analysis of the results of the MultiWOZ experiments reveals the patterns of post-processing by PPNs that contribute to the overall dialogue performance of the system.

A multi-agent collaborative algorithm for task-oriented dialogue systems

A multi-agent collaborative algorithm for task-oriented dialogue systems

Enhancing Clinical Accuracy of Medical Chatbots with Large Language Models.

The rapid advancement of large language models (LLMs) has opened up new possibilities for transforming healthcare practices, patient interactions, and medical report generation. This paper explores the application of LLMs in developing medical chatbots and virtual assistants that prioritize clinical accuracy. We propose a novel multi-turn dialogue model, including adjusting the position of layer normalization to improve training stability and convergence, employing a contextual sliding window reply prediction task to capture fine-grained local context, and developing a local critical information distillation mechanism to extract and emphasize the most relevant information. These components are integrated into a multi-turn dialogue model that generates coherent and clinically accurate responses. Experiments on the MIMIC-III and n2c2 datasets demonstrate the superiority of the proposed model over state-of-the-art baselines, achieving significant improvements in perplexity, BLEU-2, recall at K scores, medical entity recognition, and response coherence. The proposed model represents a significant step in developing reliable and contextually relevant multi-turn medical dialogue systems that can assist patients and healthcare professionals.

Mitigating the negative impact of over-association for conversational query production

Conversational query generation aims at producing search queries from dialogue histories, which are then used to retrieve relevant knowledge from a search engine to help knowledge-based dialogue systems. Trained to maximize the likelihood of gold queries, previous models suffer from the data hunger issue, and they tend to both drop important concepts from dialogue histories and generate irrelevant concepts at inference time. We attribute these issues to the over-association phenomenon where a large number of gold queries are indirectly related to the dialogue topics, because annotators may unconsciously perform reasoning with their background knowledge when generating these gold queries. We carefully analyze the negative effects of this phenomenon on pretrained Seq2seq query producers and then propose effective instance-level weighting strategies for training to mitigate these issues from multiple perspectives. Experiments on two benchmarks, Wizard-of-Internet and DuSinc, show that our strategies effectively alleviate the negative effects and lead to significant performance gains (2%∼5% across automatic metrics and human evaluation). Further analysis shows that our model selects better concepts from dialogue histories and is 10 times more data efficient than the baseline.