Artificial Intelligence (AI) is increasingly helping people with all kinds of tasks, due to its promising capabilities. In some tasks, an AI system by itself will take over tasks, but in other tasks, an AI system making decisions on its own would be undesired due to ethical and legal reasons. In those cases, AI can still be of help by forming human-AI teams, in which humans get advice from the AI system helping them with making their final decisions. Human-AI teams are for instance used in the medical and legal fields. One problem arises, in which instances should one trust an AI system and in which not? Trusting the AI system when it is correct and trusting yourself when you are correct, results in a high appropriate reliance. If users appropriately rely on AI systems, it is possible to achieve complementary team performance, which is better than any single teammate. However, as known from previous literature, people struggle with assessing their performance and knowing how well they perform compared to peers. When one overestimates their performance this can be because of a dual burden, due to the lack of skill they also lack the skill to accurately estimate their performance. This phenomenon is called the Dunning-Kruger Effect (DKE). This raises questions about whether the inability to estimate their own capabilities would also reflect on their assessment of the AI system its performance. In this thesis we look at how the DKE affects (appropriate) reliance on AI systems and if so, how such effects due to the DKE can be mitigated. The effects of the DKE and possible mitigation are being tested through an empirical study (N = 249). The attempt at mitigation is done by including a tutorial intervention, which has been proved in previous research to be useful in decreasing the DKE. The tutorial intervention is aimed at revealing the weaknesses of the participant and making them aware of their miscalibrated self-estimation. Furthermore, in this thesis, the effects of revealing how the AI system makes its decisions through explainable AI (XAI) are explored. The XAI consisted of highlights from logic unit-based explanations, it should allow participants to gain more understanding of the AI advice. This thesis shows how this will affect user self-assessment and reliance on the AI system. We found that participants who overestimate themselves tend to rely less on the AI system, compared to participants that had an accurate or underestimation of their performance. After the tutorial participants have a better calibration of their self-assessment. While the designed tutorial intervention can help participants calibrate their self-assessment, it fails to promote (appropriate) reliance. Furthermore, the logic units-based explanations did not improve accurate self-assessing or increase user (appropriate) reliance on AI systems. This thesis shows the importance of considering cognitive biases when dealing with human-AI teams and invites more research on how to handle and mitigate the DKE in human-AI decision making.

While most large language models (LLMs) are powerful, they are primarily designed for general purposes. Consequently, many enterprises and institutions have now focused on developing domain-specific models. In the realm of education, an expert LLM can significantly enhance students' ability to find information more effectively and reach their learning goals. Nevertheless, the training of such expert models in education remains largely unexplored. This study explores this research gap by developing a framework to transform semi-structured educational web data into structured datasets and perform instruction tuning on foundation models. Additionally, we conduct a comprehensive performance analysis to determine how various training factors affect model performance. We first employed a systematic and cost-effective approach involving web data extraction, data cleaning, validation, task design based on student surveys, and automated instruction instance generation using LLMs. Human evaluations confirmed the quality, especially the relevance and accuracy of these datasets. This study then investigates the impact of various training techniques on domain-specific educational large language models (LLMs) performance. Our experiments reveal that further pre-training enhances model performance, especially with domain-specific terminology, although the performance gains decrease as the dataset size increases. Furthermore, multi-task training also improves model relevance, accuracy, and clarity, but less correlated tasks and datasets can present challenges. These challenges include increased complexity and potential degradation in performance due to the model having to switch between diverse tasks. Lastly, this study conducts a comparative analysis of different models and it highlights trade-offs between computational resources and performance. The findings demonstrate that a structured approach to dataset generation and strategic training can effectively develop domain-specific LLMs in education. This research benefits the development of educational LLMs and provides a foundation for future researchers to build more specialized models in various domains.

While most large language models (LLMs) are powerful, they are primarily designed for general purposes. Consequently, many enterprises and institutions have now focused on developing domain-specific models. In the realm of education, an expert LLM can significantly enhance students' ability to find information more effectively and reach their learning goals. Nevertheless, the training of such expert models in education remains largely unexplored. This study explores this research gap by developing a framework to transform semi-structured educational web data into structured datasets and perform instruction tuning on foundation models. Additionally, we conduct a comprehensive performance analysis to determine how various training factors affect model performance. We first employed a systematic and cost-effective approach involving web data extraction, data cleaning, validation, task design based on student surveys, and automated instruction instance generation using LLMs. Human evaluations confirmed the quality, especially the relevance and accuracy of these datasets. This study then investigates the impact of various training techniques on domain-specific educational large language models (LLMs) performance. Our experiments reveal that further pre-training enhances model performance, especially with domain-specific terminology, although the performance gains decrease as the dataset size increases. Furthermore, multi-task training also improves model relevance, accuracy, and clarity, but less correlated tasks and datasets can present challenges. These challenges include increased complexity and potential degradation in performance due to the model having to switch between diverse tasks. Lastly, this study conducts a comparative analysis of different models and it highlights trade-offs between computational resources and performance. The findings demonstrate that a structured approach to dataset generation and strategic training can effectively develop domain-specific LLMs in education. This research benefits the development of educational LLMs and provides a foundation for future researchers to build more specialized models in various domains.

Common sense is knowledge that most humans have, but machines do not. Generally, computer knowledge bases make use of positive (known) knowledge. However, in addition to positive common sense knowledge, there is also negative. Negative knowledge represent facts that are known to be untrue, like "a cat does not have fins". This knowledge is important in order for a machine to make assumptions the same way a human does. This research proposes ways to combine and organize the positive and negative knowledge tuples in a unified way. The two main ways that are looked into are table-like and graph-like organizations. These are analyzed based on different requirements and important queries are defined. Based on the requirements a recommendation is made. The research also takes a look at a solution that makes use of complex number space and suggest how this solution could be further researched and improved in the future.

Commonsense knowledge is the key of human intelligence in generalizing their knowledge to deal with complex tasks. Over the past years, a lot of research has been done in both natural language processing (NLP) and computer vision (CV) on leveraging commonsense knowledge to improve AI models. However, no systematic comparisons of existing work have been made between the two domains. Therefore this survey aims to provide an overview of how commonsense knowledge is used within NLP and CV and how research varies between these two domains and what future challenges it may hold. An observation made from this survey is that leveraging commonsense is more difficult in CV than NLP, as commonsense is mostly incorporated textually and datasets need to be filtered to make them more relevant for visual commonsense. We hope to promote further research and create a better understanding of commonsense knowledge and its applications with this survey.

Commonsense knowledge plays a key role in human intelligence. It is knowledge possessed by most humans that helps them in everyday situations. One possible way is to store the knowledge in four types. Each piece is either positive or negative, and generative or discriminative. For efficient retrieval and storage, a uniform model is needed. Existing models for commonsense knowledge are not fit for negative and discriminative knowledge. The aim of this paper is to create a uniform model to store both positive and negative generative and discriminative knowledge tuples. Models are evaluated on a set of generalized queries as well as on the storage they require. Four possible models were evaluated of which two were the most promising: the generative model and the combined model. The generative model is efficient in storage and retrieving generative knowledge for concepts, but relatively slow in distinguishing concepts. Combining the generative model with discriminative tuples gives the combined model, a model that is the most efficient for all queries but expensive in storage. Which of the two models is most suitable depends on the application and the available resources.

Commonsense knowledge based question answer- ing is a recent topic that has seen a surge in inter- est. Yet most models obtain general data, this pa- per looks at obtaining query-specific similar con- cepts using first and second-order proximity to- gether with BERT-based retrieval. Using these query-specific concepts new commonsense knowl- edge can be obtained using a Game with a pur- pose. Results show that this current implementa- tion leaves room for improvement.

Powerful predictive AI systems have demonstrated great potential in augmenting human decision-making. Recent empirical work has argued that the vision for optimal human-AI collaboration requires ‘appropriate reliance’ of humans on AI systems. However, accurately estimating the trustworthiness of AI advice at the instance level is quite challenging, especially in the absence of performance feedback pertaining to the AI system. In practice, the performance disparity of machine learning models on out-of-distribution data makes the dataset-specific performance feedback unreliable in human-AI collaboration. Inspired by existing literature on critical thinking and explanation-based human debugging, we propose the use of debugging an AI system as an intervention to foster appropriate reliance. In this paper, we explore whether a critical evaluation of AI performance within a debugging setting can better calibrate users’ assessment of an AI system and lead to more appropriate reliance. Through a quantitative empirical study (N = 234), we found that our proposed debugging intervention does not work as expected in facilitating appropriate reliance. Instead, we observe a decrease in reliance on the AI system after the intervention — potentially resulting from early exposure to the AI system’s weakness. We explored the dynamics of user confidence to help explain how inappropriate reliance patterns occur and found that human confidence is not independent of AI advice, which is potentially dangerous when trying to achieve appropriate reliance. Our findings have important implications for designing effective interventions to facilitate appropriate reliance and better human-AI collaboration

Commonsense knowledge is information that all humans own and use to interpret common situations and react to them accordingly. This kind of information is necessary for the training of artificial intelligence models to reach a performance as close as possible to human performance. Researchers have developed methods that use crowdsourcing to collect this kind of knowledge from the general public. This research focuses on systematically surveying the existing literature about these methods. We created a taxonomy to describe and compare the existing work based on the following three measures that were the most common ones reported: efficiency, cost, and quality.

Commonsense knowledge is a type of knowledge consisting of facts that humans use every day. Humans make queries in search engines with different user intents, and some of them can be answered by knowledge tuples. Different types of knowledge are stored differently in the knowledge bases. Being aware of the types of commonsense knowledge required to answer the queries can accelerate the process of finding corresponding knowledge for the search engines to give a response to users. For some queries with specific user intents, it is not possible to be answered solely with commonsense knowledge because some analysis and judgment from humans are needed. On the other hand, some queries can be answered with commonsense knowledge tuples and the user intents can have a strong indication of what the knowledge type is required to answer. The research is to look into how to map queries and their user intents to knowledge types and explore the impacts of user intents in the knowledge type classification. There was no existing dataset that had annotations on both user intents and knowledge types. In this work, the described dataset was created. Observations of the created dataset and experiments on three classifiers with accuracy being around 0.99 were conducted. The results show that user intents generally help the classification of the type of commonsense knowledge.

Common sense knowledge (CSK) comes naturally to humans, but is very hard for computers to comprehend. However it is critical for machines to behave intelligently, and as such collecting CSK has become a prevalent field of research. Whilst a lot of research has been done to develop CSK acquisition methods, not much work has been done to survey the literature that already exists. Furthermore the surveys that have been done are outdated, and as such there is a clear gap in the literature. This paper will survey the different approaches to CSK acquisition and evaluate their effectiveness, as a way of gauging their real life applicability. It will also compare the current extit{state of the art} methods, to some previous work to illustrate the progress that has been made and project that into the future. Furthermore this paper will also create a taxonomy categorizing the surveyed literature in order give a better overview of existing methods. Finally, from the literature surveyed it is clear that these methods have made a lot of progress, but aren't quite yet at the same level as human performance. Nevertheless they have become robust enough to be deployed in real applications.

Trust is a fundamental component in human-AI relationships, serving as a critical element of user acceptance and satisfaction, particularly within the realm of Decision Support Systems (DSS). The technological advances in conversational user interfaces (CUIs) such as ChatGPT and digital assistants (e.g., Alexa) allow laypeople to interact with DSS without knowing the mechanisms behind them. Extensive research has explored the benefits of CUIs and strives to improve their usability and adoption rate. However, while interacting with such CUIs, how to facilitate proper user trust for decision support is still under-explored. To address the research gap, we aim to test the impact of emotional expressiveness in CUIs on building user trust. To analyze the impact of emotional expressiveness in CUIs to build user trust and whether voice-based CUI is more efficient in building user trust compared to text-based CUI. We implemented a conversational interface with varying emotional expressiveness that can serve six conditions: two text-based and four voice-based. Text-based CUIs are differentiated by lexical expressiveness. Voice-based CUIs are varying in both lexical expressiveness and prosodic expressiveness. Regardless of the modality and emotional expressiveness, each CUI serves as an interactive medium for users with the DSS, which supports them to find a suitable house given a scenario. Through an empirical study (N = 151), the experimental results are insufficient to conclude the impact of prosodic expressiveness and lexical expressiveness on user trust and usability in CUIs. In addition, we did not find any statistically significant difference between text-based and voice-based CUIs in trust or perceived usability. Our findings can potentially be explained by the uncanniness effect [46]: initially, increased emotional expressiveness in a chatbot could positively influence user trust, but over time this could turn into a negative impact. These results offer a potential way to explain the complex dynamics of trust in conversational DSS and some implications in chatbot design within the context of DSS. Our findings can benefit the future design and development of conversational agents-based DSS by considering emotional expressiveness.