Natural language processing (NLP) has witnessed a profound impact of large language models (LLMs) that excel in a multitude of tasks. However, the limitation of LLMs in multilingual settings, particularly in underrepresented languages, remains a significant hurdle. This thesis aims to bridge the gap in NLP research and development by focusing on underrepresented languages. A comprehensive evaluation of LLMs is conducted to assess their capabilities in these languages, revealing the challenges of multilingual and multicultural generalization. Addressing the multilingual generalization gap, this thesis proposes data-and-compute-efficient methods to mitigate the disparity in LLM ability in underrepresented languages, allowing better generalization on underrepresented languages without the loss of task generalization ability. The proposed solutions cover cross-lingual continual instruction tuning, retrieval-based cross-lingual in-context learning, and in-context query alignment. Furthermore, a novel method to measure cultural values alignment between LLMs operating in different languages is proposed, ensuring cultural sensitivity and inclusivity. These contributions aim to enhance the multilingual and multicultural alignment of LLMs in underrepresented languages, ultimately advancing the NLP field toward greater equality and inclusiveness.

Service robots in human-centered environments such as hospitals, office buildings, and long-term care homes need to navigate while adhering to social norms to ensure the safety and comfortability of the people they are sharing the space with. Furthermore, they need to adapt to new social scenarios that can arise during robot navigation. In this paper, we present a novel Online Lifelong Vision Language architecture, OLiVia-Nav, which uniquely integrates vision-language models (VLMs) with an online lifelong learning framework for robot social navigation. We introduce a unique distillation approach, Social Context Contrastive Language Image Pre-training (SC-CLIP), to transfer the social reasoning capabilities of large VLMs to a lightweight VLM, in order for OLiVia-Nav to directly encode social and environment context during robot navigation. These encoded embeddings are used to generate and select robot social compliant trajectories. The lifelong learning capabilities of SC-CLIP enable OLiVia-Nav to update the lightweight VLM with robot trajectory predictions overtime as new social scenarios are encountered. We conducted extensive real-world experiments in diverse social navigation scenarios. The results showed that OLiVia-Nav outperformed existing state-of-the-art DRL and VLM methods in terms of mean squared error, Hausdorff loss, and personal space violation duration. Ablation studies also verified the design choices for OLiVia-Nav.

Emotion cognition in large language models (LLMs) is crucial for enhancing performance across various applications, such as social media, human-computer interaction, and mental health assessment. We explore the current landscape of research, which primarily revolves around emotion classification, emotionally rich response generation, and Theory of Mind assessments, while acknowledge the challenges like dependency on annotated data and complexity in emotion processing. In this paper, we present a detailed survey of recent progress in LLMs for emotion cognition. We explore key research studies, methodologies, outcomes, and resources, aligning them with Ulric Neisser's cognitive stages. Additionally, we outline potential future directions for research in this evolving field, including unsupervised learning approaches and the development of more complex and interpretable emotion cognition LLMs. We also discuss advanced methods such as contrastive learning used to improve LLMs' emotion cognition capabilities.

The text below arose from a course on'Mathematical Data Science' that I taught twice for final year BSc Mathematics students in the UK between 2019 and 2020. The notes presently cover the first part (roughly a third) of the course focussing on the characteristics and peculiarities of high-dimensional data. An improved version of the notes appeared as part of the textbook [7]; we refer the reader in particular to [7, Chapters 8 -12]. I would like to thank my former students who attended the course and helped me with their feedback to write these lecture notes. Concrete examples are as follows. Each user can give a rating from one to five stars for each movie. When doing medical diagnostic tests, we can represent a subject by the vector containing her/his results. These can include integers like antibody counts, real numbers like temperature, pairs of real numbers like blood pressure, or binary values like if a subject has tested positive or negative for a certain infection. If we name the users 1, 2, 3,..., we can represent user j in R Given such a high-dimensional data set A, classical tasks to analyze the data, or make predictions based on it, involve to compute distances between data points. This can be for example the classical euclidean distance (or any other p-norm), CHAPTER 1. THE CURSE OF HIGH DIMENSIONS 4 However, if d is very large, we are faced with the following two obstructions.

While each of the lower-resource scenarios bears its unique socio-historical contexts, the tutorial (Selected as a tutorial at COLING 2025) brings together researchers working separately in Despite excellent results on benchmarks these scenarios. Collectively, the tutorial will connect over a small subset of languages, large language past research in terms of: models struggle to process text from Challenges in data curation languages situated in'lower-resource' scenarios Potential for wide linguistic variation (e.g., existing such as dialects/sociolects (national on a linguistic continuum or eschewing or social varieties of a language), Creoles strict spelling conventions, etc.) (languages arising from linguistic contact Need for smart modeling choices over greedy between multiple languages) and other lowresource ones languages. This introductory Increased model vulnerability tutorial will identify common challenges, This introductory tutorial identifies the emergence approaches, and themes in natural language of'lower-resource' scenarios, specifically national processing (NLP) research for confronting varieties, Creoles and other low-resource languages, and overcoming the obstacles inherent and highlights commonalities and differences to data poor contexts.

Figure 1: A human user interacting with a virtual UR5e robot to provide a trajectory demonstration as the robot carries a coffee mug over a table with a laptop on top. Extended reality provides a natural setting for demonstrating robotic trajectories while bypassing safety Many real-world tasks are intuitive for a human to perform, but difficult concerns and providing a broader range of interaction modalities. In these scenarios, robotic systems can benefit from expert (RADER) system, a generic extended reality interface for learning demonstrations, wherein human operators physically move the from demonstration. We additionally present its application to an robot along trajectories, to learn how to perform each task. In many existing state-of-the-art learning from demonstration approach and settings, it may be difficult or unsafe to use a physical robot to provide show comparable results between demonstrations given on a physical these demonstrations, for example, considering cooking tasks robot and those given using our extended reality system.

With the recent evolution of large language models (LLMs), concerns about aligning such models with human values have grown. Previous research has primarily focused on assessing LLMs' performance in terms of the Helpful, Honest, Harmless (3H) basic principles, while often overlooking their alignment with educational values in the Chinese context. To fill this gap, we present Edu-Values, the first Chinese education values evaluation benchmark designed to measure LLMs' alignment ability across seven dimensions: professional ideology, cultural literacy, educational knowledge and skills, education laws and regulations, teachers' professional ethics, basic competencies, and subject knowledge. We meticulously design and compile 1,418 questions, including multiple-choice, multi-modal question answering, subjective analysis, adversarial prompts, and questions on traditional Chinese culture. We conduct both human evaluation and automatic evaluation over 11 state-of-the-art (SoTA) LLMs, and highlight three main findings: (1) due to differences in educational culture, Chinese LLMs significantly outperform English LLMs, with Qwen 2 ranking the first with a score of 81.37; (2) LLMs perform well in subject knowledge and teaching skills but struggle with teachers' professional ethics and basic competencies; (3) LLMs excel at multiple-choice questions but perform poorly on subjective analysis and multi-modal tasks. This demonstrates the effectiveness and potential of the proposed benchmark. Our dataset is available at https://github.com/zhangpeii/Edu-Values.git.

Robust policies enable reinforcement learning agents to effectively adapt to and operate in unpredictable, dynamic, and ever-changing real-world environments. Factored representations, which break down complex state and action spaces into distinct components, can improve generalization and sample efficiency in policy learning. In this paper, we explore how the curriculum of an agent using a factored state representation affects the robustness of the learned policy. We experimentally demonstrate three simple curricula, such as varying only the variable of highest regret between episodes, that can significantly enhance policy robustness, offering practical insights for reinforcement learning in complex environments.

Graph signal processing deals with algorithms and signal representations that leverage graph structures for multivariate data analysis. Often said graph topology is not readily available and may be time-varying, hence (dynamic) graph structure learning from nodal (e.g., sensor) observations becomes a critical first step. In this paper, we develop a novel algorithm for online graph learning using observation streams, assumed to be smooth on the latent graph. Unlike batch algorithms for topology identification from smooth signals, our modus operandi is to process graph signals sequentially and thus keep memory and computational costs in check. To solve the resulting smoothness-regularized, time-varying inverse problem, we develop online and lightweight iterations built upon the proximal variant of the alternating direction method of multipliers (ADMM), well known for its fast convergence in batch settings. The proximal term in the topology updates seamlessly implements a temporal-variation regularization, and we argue the online procedure exhibits sublinear static regret under some simplifying assumptions. Reproducible experiments with synthetic and real graphs demonstrate the effectiveness of our method in adapting to streaming signals and tracking slowly-varying network connectivity. The proposed approach also exhibits better tracking performance (in terms of suboptimality), when compared to state-of-the-art online graph learning baselines.

Adversarial attacks exploit vulnerabilities in a model's decision boundaries through small, carefully crafted perturbations that lead to significant mispredictions. In 3D vision, the high dimensionality and sparsity of data greatly expand the attack surface, making 3D vision particularly vulnerable for safety-critical robotics. To enhance 3D vision's adversarial robustness, we propose a training objective that simultaneously minimizes prediction loss and mutual information (MI) under adversarial perturbations to contain the upper bound of misprediction errors. This approach simplifies handling adversarial examples compared to conventional methods, which require explicit searching and training on adversarial samples. However, minimizing prediction loss conflicts with minimizing MI, leading to reduced robustness and catastrophic forgetting. To address this, we integrate curriculum advisors in the training setup that gradually introduce adversarial objectives to balance training and prevent models from being overwhelmed by difficult cases early in the process. The advisors also enhance robustness by encouraging training on diverse MI examples through entropy regularizers. We evaluated our method on ModelNet40 and KITTI using PointNet, DGCNN, SECOND, and PointTransformers, achieving 2-5% accuracy gains on ModelNet40 and a 5-10% mAP improvement in object detection. Our code is publicly available at https://github.com/nstrndrbi/Mine-N-Learn.