Detecting collaborative and problem-solving behaviours from digital traces to interpret students' collaborative problem solving (CPS) competency is a long-term goal in the Artificial Intelligence in Education (AIEd) field. Although multimodal data and advanced models are argued to have the potential to detect complex CPS behaviours, empirical evidence on their value remains limited with some contrasting evidence. In this study, we investigated the potential of multimodal data to improve model performance in diagnosing 78 secondary school students' CPS subskills and indicators in authentic educational settings. In particular, text embeddings from verbal data and acoustic embed-dings from audio data were used in a multimodal classification model for CPS diagnosis. Both unimodal and multimodal transformer-based models outperformed traditional models in detecting CPS classes. Although the inclusion of multimodality did not improve the performance of traditional unimodal models, its integration into transformer-based models demonstrated improved performance for diagnosing social-cognitive CPS classes compared to unimodal transformer-based models. Based on the results, the paper argues that multimodality and the selection of a particular modelling technique should not be taken for granted to achieve the best performance in the automated detection of every CPS subskill and indicator. Rather, their value is limited to certain types of CPS indicators, affected by the complexity of the labels, and dependent on the composition of indicators in the dataset. We conclude the paper by discussing the required nuance when considering the value of LLMs and multimodality in automated CPS diagnosis, highlighting the need for human-AI complementarity, and proposing the exploration of relevant model architectures and techniques to improve CPS diagnosis in authentic educational contexts.

While hallucinations of large language models could been alleviated through retrieval-augmented generation and citation generation, how the model utilizes internal knowledge is still opaque, and the trustworthiness of its generated answers remains questionable. In this work, we introduce Context-Prior Augmented Citation Generation task, requiring models to generate citations considering both external and internal knowledge while providing trustworthy references, with 5 evaluation metrics focusing on 3 aspects: answer helpfulness, citation faithfulness, and trustworthiness. We introduce RAEL, the paradigm for our task, and also design INTRALIGN, an integrated method containing customary data generation and an alignment algorithm. Our experimental results show that our method achieves a better cross-scenario performance with regard to other baselines. Our extended experiments further reveal that retrieval quality, question types, and model knowledge have considerable influence on the trustworthiness in citation generation.

Vision-language models (VLMs) excel in various multimodal tasks but frequently suffer from poor calibration, resulting in misalignment between their verbalized confidence and response correctness. This miscalibration undermines user trust, especially when models confidently provide incorrect or fabricated information. In this work, we propose a novel Confidence Calibration through Semantic Perturbation (CSP) framework to improve the calibration of verbalized confidence for VLMs in response to object-centric queries. We first introduce a perturbed dataset where Gaussian noise is applied to the key object regions to simulate visual uncertainty at different confidence levels, establishing an explicit mapping between visual ambiguity and confidence levels. We further enhance calibration through a two-stage training process combining supervised fine-tuning on the perturbed dataset with subsequent preference optimization. Extensive experiments on popular benchmarks demonstrate that our method significantly improves the alignment between verbalized confidence and response correctness while maintaining or enhancing overall task performance. These results highlight the potential of semantic perturbation as a practical tool for improving the reliability and interpretability of VLMs.

The small sample imbalance (S&I) problem is a major challenge in machine learning and data analysis. It is characterized by a small number of samples and an imbalanced class distribution, which leads to poor model performance. In addition, indistinct inter-class feature distributions further complicate classification tasks. Existing methods often rely on algorithmic heuristics without sufficiently analyzing the underlying data characteristics. We argue that a detailed analysis from the data perspective is essential before developing an appropriate solution. Therefore, this paper proposes a systematic analytical framework for the S\&I problem. We first summarize imbalance metrics and complexity analysis methods, highlighting the need for interpretable benchmarks to characterize S&I problems. Second, we review recent solutions for conventional, complexity-based, and extreme S&I problems, revealing methodological differences in handling various data distributions. Our summary finds that resampling remains a widely adopted solution. However, we conduct experiments on binary and multiclass datasets, revealing that classifier performance differences significantly exceed the improvements achieved through resampling. Finally, this paper highlights open questions and discusses future trends.

In the field of natural language processing (NLP), the capability to infer and reason about causality is increasingly recognized as a critical component of intelligent systems. Despite the recent advancement of large language models (LLMs) (Radford et al., 2019; Devlin et al., 2019; Brown et al., 2020; Zhang et al., 2022; OpenAI, 2023; Ignat et al., 2024, inter alia), a key question still remains: Can these models understand and reason about causality? This is a critical skill before we can trust AI agents to be integrated into decision-making systems. Moreover, even if LLMs succeed at some extent of reasoning, they still lack transparency of how their decisions are made, forming a strong need for interpretabil-ity (Luo and Specia, 2024; Räuker et al., 2023; Zou et al., 2023). T o bridge the gap, this thesis explores various facets of causal reasoning in LLMs. W e present a series of studies that collectively advance the knowledge of how well these models perform causal reasoning (Part I), how their decisions are made (Part II), how causality among learning variables influences NLP tasks (Part III), and how causality and NLP can together analyze social problems (Part IV). Below we introduce an overview of the four parts and their corresponding chapters.

The rapid growth of healthcare data and advances in computational power have accelerated the adoption of artificial intelligence (AI) in medicine. However, AI systems deployed without explicit fairness considerations risk exacerbating existing healthcare disparities, potentially leading to inequitable resource allocation and diagnostic disparities across demographic subgroups. To address this challenge, we propose FairGrad, a novel gradient reconciliation framework that automatically balances predictive performance and multi-attribute fairness optimization in healthcare AI models. Our method resolves conflicting optimization objectives by projecting each gradient vector onto the orthogonal plane of the others, thereby regularizing the optimization trajectory to ensure equitable consideration of all objectives. Evaluated on diverse real-world healthcare datasets and predictive tasks - including Substance Use Disorder (SUD) treatment and sepsis mortality - FairGrad achieved statistically significant improvements in multi-attribute fairness metrics (e.g., equalized odds) while maintaining competitive predictive accuracy. These results demonstrate the viability of harmonizing fairness and utility in mission-critical medical AI applications.

The requirement for autonomous robots to exhibit higher-level cognitive skills by planning and adapting in an ever-changing environment is indeed a great challenge for the AI community. Progress has been made in the automated planning community on refinement and repair of an agent's symbolic knowledge to do task planning in an incomplete or changing environmental model, but these advances up to now have not been transferred to real physical robots. This paper demonstrates how a physical robot can be capable of adapting its symbolic knowledge of the environment, by using experiences in robot action execution to drive knowledge refinement and hence to improve the success rate of the task plans the robot creates. To implement more robust planning systems, we propose a method for refining domain knowledge to improve the knowledge on which intelligent robot behavior is based. This architecture has been implemented and evaluated using a NAO robot. The refined knowledge leads to the future synthesis of task plans which demonstrate decreasing rates of failure over time as faulty knowledge is removed or adjusted.

The undergoing energy transition is causing behavioral changes in electricity use, e.g. with self-consumption of local generation, or flexibility services for demand control. To better understand these changes and the challenges they induce, accessing individual smart meter data is crucial. Yet this is personal data under the European GDPR. A widespread use of such data requires thus to create synthetic realistic and privacy-preserving samples. This paper introduces a new synthetic load curve dataset generated by conditional latent diffusion. We also provide the contracted power, time-of-use plan and local temperature used for generation. Fidelity, utility and privacy of the dataset are thoroughly evaluated, demonstrating its good quality and thereby supporting its interest for energy modeling applications.

Our study introduces an automated system leveraging large language models (LLMs) to assess the effectiveness of five key tutoring strategies: 1. giving effective praise, 2. reacting to errors, 3. determining what students know, 4. helping students manage inequity, and 5. responding to negative self-talk. Using a public dataset from the Teacher-Student Chatroom Corpus, our system classifies each tutoring strategy as either being employed as desired or undesired. Our study utilizes GPT-3.5 with few-shot prompting to assess the use of these strategies and analyze tutoring dialogues. The results show that for the five tutoring strategies, True Negative Rates (TNR) range from 0.655 to 0.738, and Recall ranges from 0.327 to 0.432, indicating that the model is effective at excluding incorrect classifications but struggles to consistently identify the correct strategy. The strategy \textit{helping students manage inequity} showed the highest performance with a TNR of 0.738 and Recall of 0.432. The study highlights the potential of LLMs in tutoring strategy analysis and outlines directions for future improvements, including incorporating more advanced models for more nuanced feedback.

Constrained optimization with multiple functional inequality constraints has significant applications in machine learning. This paper examines a crucial subset of such problems where both the objective and constraint functions are weakly convex. Existing methods often face limitations, including slow convergence rates or reliance on double-loop algorithmic designs. To overcome these challenges, we introduce a novel single-loop penalty-based stochastic algorithm. Following the classical exact penalty method, our approach employs a {\bf hinge-based penalty}, which permits the use of a constant penalty parameter, enabling us to achieve a {\bf state-of-the-art complexity} for finding an approximate Karush-Kuhn-Tucker (KKT) solution. We further extend our algorithm to address finite-sum coupled compositional objectives, which are prevalent in artificial intelligence applications, establishing improved complexity over existing approaches. Finally, we validate our method through experiments on fair learning with receiver operating characteristic (ROC) fairness constraints and continual learning with non-forgetting constraints.