Language models (LMs) trained on vast quantities of text data can acquire sophisticated skills such as generating summaries, answering questions or generating code. However, they also manifest behaviors that violate human preferences, e.g., they can generate offensive content, falsehoods or perpetuate social biases. In this thesis, I explore several approaches to aligning LMs with human preferences. First, I argue that aligning LMs can be seen as Bayesian inference: conditioning a prior (base, pretrained LM) on evidence about human preferences (Chapter 2). Conditioning on human preferences can be implemented in numerous ways. In Chapter 3, I investigate the relation between two approaches to finetuning pretrained LMs using feedback given by a scoring function: reinforcement learning from human feedback (RLHF) and distribution matching. I show that RLHF can be seen as a special case of distribution matching but distributional matching is strictly more general. In chapter 4, I show how to extend the distribution matching to conditional language models. Finally, in chapter 5 I explore a different root: conditioning an LM on human preferences already during pretraining. I show that involving human feedback from the very start tends to be more effective than using it only during supervised finetuning. Overall, these results highlight the room for alignment techniques different from and complementary to RLHF.

This research paper will bring forth the innovative pedagogical approach in computer science education, which uses a combination of methodologies borrowed from Artificial Intelligence (AI) and DevOps to enhance the learning experience in Content Management Systems (CMS) Development. It has been done over three academic years, comparing the traditional way of teaching with the lately introduced AI-supported techniques. This had three structured sprints, each one of them covering the major parts of the sprint: object-oriented PHP, theme development, and plugin development. In each sprint, the student deals with part of the theoretical content and part of the practical task, using ChatGPT as an auxiliary tool. In that sprint, the model will provide solutions in code debugging and extensions of complex problems. The course includes practical examples like code replication with PHP, functionality expansion of the CMS, even development of custom plugins, and themes. The course practice includes versions' control with Git repositories. Efficiency will touch the theme and plugin output rates during development and mobile/web application development. Comparative analysis indicates that there is a marked increase in efficiency and shows effectiveness with the proposed AI- and DevOps-supported methodology. The study is very informative since education in computer science and its landscape change embodies an emerging technology that could have transformation impacts on amplifying the potential for scalable and adaptive learning approaches.

Additionally, it explores how LLMs can be more closely aligned with human preferences through innovative reinforcement learning frameworks and other novel methods that incorporate human feedback. The article also examines the emerging technique of retrieval augmented generation, integrating external knowledge into LLMs. The ethical dimensions of LLM deployment are discussed, underscoring the need for mindful and responsible application. Concluding with a perspective on future research trajectories, this review offers a succinct yet comprehensive overview of the current state and emerging trends in the evolving landscape of LLMs, serving as an insightful guide for both researchers and practitioners in artificial intelligence.

This paper introduces a novel approach, evolutionary multi-objective optimisation for fairness-aware self-adjusting memory classifiers, designed to enhance fairness in machine learning algorithms applied to data stream classification. With the growing concern over discrimination in algorithmic decision-making, particularly in dynamic data stream environments, there is a need for methods that ensure fair treatment of individuals across sensitive attributes like race or gender. The proposed approach addresses this challenge by integrating the strengths of the self-adjusting memory K-Nearest-Neighbour algorithm with evolutionary multi-objective optimisation. This combination allows the new approach to efficiently manage concept drift in streaming data and leverage the flexibility of evolutionary multi-objective optimisation to maximise accuracy and minimise discrimination simultaneously. We demonstrate the effectiveness of the proposed approach through extensive experiments on various datasets, comparing its performance against several baseline methods in terms of accuracy and fairness metrics. Our results show that the proposed approach maintains competitive accuracy and significantly reduces discrimination, highlighting its potential as a robust solution for fairness-aware data stream classification. Further analyses also confirm the effectiveness of the strategies to trigger evolutionary multi-objective optimisation and adapt classifiers in the proposed approach.

The Socratic method is a way of guiding students toward solving a problem independently without directly revealing the solution to the problem. Although this method has been shown to significantly improve student learning outcomes, it remains a complex labor-intensive task for instructors. Large language models (LLMs) can be used to augment human effort by automatically generating Socratic questions for students. However, existing methods that involve prompting these LLMs sometimes produce invalid outputs, e.g., those that directly reveal the solution to the problem or provide irrelevant or premature questions. To alleviate this problem, inspired by reinforcement learning with AI feedback (RLAIF), we first propose a data augmentation method to enrich existing Socratic questioning datasets with questions that are invalid in specific ways. Next, we propose a method to optimize open-source LLMs such as LLama 2 to prefer ground-truth questions over generated invalid ones, using direct preference optimization (DPO). Our experiments on a Socratic questions dataset for student code debugging show that a DPO-optimized 7B LLama 2 model can effectively avoid generating invalid questions, and as a result, outperforms existing state-of-the-art prompting methods.

The advent of generative artificial intelligence (GenAI) technologies has revolutionized research, with significant implications for Digital Humanities (DH), a field inherently intertwined with technological progress. This article investigates how digital humanities scholars adopt, practice, as well as critically evaluate, GenAI technologies such as ChatGPT in the research process. Drawing on 76 responses collected from an international survey study, we explored digital humanities scholars' rationale for GenAI adoption in research, identified specific use cases and practices of using GenAI to support various DH research tasks, and analyzed scholars' collective perceptions of GenAI's benefits, risks, and impact on DH research. The survey results suggest that DH research communities hold divisive sentiments towards the value of GenAI in DH scholarship, whereas the actual usage diversifies among individuals and across research tasks. Our survey-based analysis has the potential to serve as a basis for further empirical research on the impact of GenAI on the evolution of DH scholarship.

Recent advances in large language models (LLM) have enabled richer social simulations, allowing for the study of various social phenomena. However, most recent work has used a more omniscient perspective on these simulations (e.g., single LLM to generate all interlocutors), which is fundamentally at odds with the non-omniscient, information asymmetric interactions that involve humans and AI agents in the real world. To examine these differences, we develop an evaluation framework to simulate social interactions with LLMs in various settings (omniscient, non-omniscient). Our experiments show that LLMs perform better in unrealistic, omniscient simulation settings but struggle in ones that more accurately reflect real-world conditions with information asymmetry. Our findings indicate that addressing information asymmetry remains a fundamental challenge for LLM-based agents.

We study both stream-based and pool-based active learning with neural network approximations. A recent line of works proposed bandit-based approaches that transformed active learning into a bandit problem, achieving both theoretical and empirical success. However, the performance and computational costs of these methods may be susceptible to the number of classes, denoted as K, due to this transformation. Therefore, this paper seeks to answer the question: "How can we mitigate the adverse impacts of K while retaining the advantages of principled exploration and provable performance guarantees in active learning?" To tackle this challenge, we propose two algorithms based on the newly designed exploitation and exploration neural networks for stream-based and pool-based active learning. Subsequently, we provide theoretical performance guarantees for both algorithms in a non-parametric setting, demonstrating a slower error-growth rate concerning K for the proposed approaches. We use extensive experiments to evaluate the proposed algorithms, which consistently outperform state-of-the-art baselines. Active learning is one of the primary areas in machine learning to investigate the learning technique on a small subset of labeled data while acquiring good generalization performance compared to passive learning [19]. There are mainly two settings of active learning: stream-based and pool-based settings.

Data analysts have long sought to turn unstructured text data into meaningful concepts. Though common, topic modeling and clustering focus on lower-level keywords and require significant interpretative work. We introduce concept induction, a computational process that instead produces high-level concepts, defined by explicit inclusion criteria, from unstructured text. For a dataset of toxic online comments, where a state-of-the-art BERTopic model outputs "women, power, female," concept induction produces high-level concepts such as "Criticism of traditional gender roles" and "Dismissal of women's concerns." We present LLooM, a concept induction algorithm that leverages large language models to iteratively synthesize sampled text and propose human-interpretable concepts of increasing generality. We then instantiate LLooM in a mixed-initiative text analysis tool, enabling analysts to shift their attention from interpreting topics to engaging in theory-driven analysis. Through technical evaluations and four analysis scenarios ranging from literature review to content moderation, we find that LLooM's concepts improve upon the prior art of topic models in terms of quality and data coverage. In expert case studies, LLooM helped researchers to uncover new insights even from familiar datasets, for example by suggesting a previously unnoticed concept of attacks on out-party stances in a political social media dataset.

Evaluation benchmarks play a pivotal role in the development of artificial intelligence (AI) systems. Traditionally, natural language processing (NLP) benchmarks have primarily focused on assessing specific and relatively straightforward abilities. However, the advent of large language models (LLMs), also known as foundation models, has brought about a paradigm shift. These powerful models have demonstrated a wide array of novel capabilities, prompting a redirection in the evaluation focus towards more general and intricate skills, such as comprehensive world knowledge and complex reasoning abilities. To align with the remarkable advancements in LLMs, new benchmarks have emerged to probe the diverse and multifaceted capabilities of these models. For instance, MMLU [8], HellaSwag [25], ARC [4], and TruthfulQA [10] are benchmark datasets that have garnered widespread recognition among researchers and are frequently employed on leaderboards to evaluate the performance of language models. However, these benchmarks are primarily tailored to the English language, resulting in a limited understanding of LLMs' capabilities in other languages, including Vietnamese. Despite the recent surge in powerful Vietnamese LLMs, such as Vistral-7B-Chat [12], PhoGPT-4B-Chat [13], and VinaLLaMA-7B-Chat [16], benchmarking these models on datasets translated from English to Vietnamese, even with perfect translations, cannot adequately assess the true quality of these language models concerning their knowledge about core interests of Vietnamese users.