In this paper, we explore the impact of augmenting pre-trained Encoder-Decoder models, specifically T5, with linguistic knowledge for the prediction of a target task. In particular, we investigate whether fine-tuning a T5 model on an intermediate task that predicts structural linguistic properties of sentences modifies its performance in the target task of predicting sentence-level complexity. Our study encompasses diverse experiments conducted on Italian and English datasets, employing both monolingual and multilingual T5 models at various sizes. Results obtained for both languages and in cross-lingual configurations show that linguistically motivated intermediate fine-tuning has generally a positive impact on target task performance, especially when applied to smaller models and in scenarios with limited data availability.

Several adaptations of Transformers models have been developed in various domains since its breakthrough in Natural Language Processing (NLP). This trend has spread into the field of Music Information Retrieval (MIR), including studies processing music data. However, the practice of leveraging NLP tools for symbolic music data is not novel in MIR. Music has been frequently compared to language, as they share several similarities, including sequential representations of text and music. These analogies are also reflected through similar tasks in MIR and NLP. This survey reviews NLP methods applied to symbolic music generation and information retrieval studies following two axes. We first propose an overview of representations of symbolic music adapted from natural language sequential representations. Such representations are designed by considering the specificities of symbolic music. These representations are then processed by models. Such models, possibly originally developed for text and adapted for symbolic music, are trained on various tasks. We describe these models, in particular deep learning models, through different prisms, highlighting music-specialized mechanisms. We finally present a discussion surrounding the effective use of NLP tools for symbolic music data. This includes technical issues regarding NLP methods and fundamental differences between text and music, which may open several doors for further research into more effectively adapting NLP tools to symbolic MIR.

As foundation models (FMs) continue to shape the landscape of AI, the in-context learning (ICL) paradigm thrives but also encounters issues such as toxicity, hallucination, disparity, adversarial vulnerability, and inconsistency. Ensuring the reliability and responsibility of FMs is crucial for the sustainable development of the AI ecosystem. In this concise overview, we investigate recent advancements in enhancing the reliability and trustworthiness of FMs within ICL frameworks, focusing on four key methodologies, each with its corresponding subgoals. We sincerely hope this paper can provide valuable insights for researchers and practitioners endeavoring to build safe and dependable FMs and foster a stable and consistent ICL environment, thereby unlocking their vast potential.

Deep Learning is the state-of-the-art technology for segmenting brain tumours. However, this requires a lot of high-quality data, which is difficult to obtain, especially in the medical field. Therefore, our solutions address this problem by using unconventional mechanisms for data augmentation. Generative adversarial networks and registration are used to massively increase the amount of available samples for training three different deep learning models for brain tumour segmentation, the first task of the BraTS2023 challenge. The first model is the standard nnU-Net, the second is the Swin UNETR and the third is the winning solution of the BraTS 2021 Challenge. The entire pipeline is built on the nnU-Net implementation, except for the generation of the synthetic data. The use of convolutional algorithms and transformers is able to fill each other's knowledge gaps. Using the new metric, our best solution achieves the dice results 0.9005, 0.8673, 0.8509 and HD95 14.940, 14.467, 17.699 (whole tumour, tumour core and enhancing tumour) in the validation set.

Social dilemmas (SDs, e.g., prisoner's dilemma), spanning various domains including environmental pollution, public health crises, and resource management, present a fundamental conflict between personal interests and the common good [Nowak, 2006]. While cooperation is beneficial for the collective, individuals are tempted to exploit or free-ride others' efforts, potentially leading to a tragedy of the commons. Historically rooted in the study of biological altruism [Smith, 1982], the traditional research on cooperation in SDs has unveiled the pivotal roles of reciprocity and social preferences in fostering cooperative behaviors in human societies [Fehr et al., 2002; Rand and Nowak, 2013]. Recently, propelled by advances in artificial intelligence (AI), this field has been undergoing a profound transformation--as AI agents now increasingly represent and engage with humans, our understanding of how cooperation emerges, evolves, and sustains in SDs is being significantly reshaped. This is particularly evident in two lines of research: multi-agent cooperation, where AI agents interact with each other in SDs, and human-agent cooperation, which examines the intricacies of human interactions with AI agents in SDs.

This paper embarks on an exploration into the Large Language Model (LLM) datasets, which play a crucial role in the remarkable advancements of LLMs. The datasets serve as the foundational infrastructure analogous to a root system that sustains and nurtures the development of LLMs. Consequently, examination of these datasets emerges as a critical topic in research. In order to address the current lack of a comprehensive overview and thorough analysis of LLM datasets, and to gain insights into their current status and future trends, this survey consolidates and categorizes the fundamental aspects of LLM datasets from five perspectives: (1) Pre-training Corpora; (2) Instruction Fine-tuning Datasets; (3) Preference Datasets; (4) Evaluation Datasets; (5) Traditional Natural Language Processing (NLP) Datasets. The survey sheds light on the prevailing challenges and points out potential avenues for future investigation. Additionally, a comprehensive review of the existing available dataset resources is also provided, including statistics from 444 datasets, covering 8 language categories and spanning 32 domains. Information from 20 dimensions is incorporated into the dataset statistics. The total data size surveyed surpasses 774.5 TB for pre-training corpora and 700M instances for other datasets. We aim to present the entire landscape of LLM text datasets, serving as a comprehensive reference for researchers in this field and contributing to future studies. Related resources are available at: https://github.com/lmmlzn/Awesome-LLMs-Datasets.

This paper proposes an innovative Attention-GAN framework for enhancing cybersecurity, focusing on anomaly detection. In response to the challenges posed by the constantly evolving nature of cyber threats, the proposed approach aims to generate diverse and realistic synthetic attack scenarios, thereby enriching the dataset and improving threat identification. Integrating attention mechanisms with Generative Adversarial Networks (GANs) is a key feature of the proposed method. The attention mechanism enhances the model's ability to focus on relevant features, essential for detecting subtle and complex attack patterns. In addition, GANs address the issue of data scarcity by generating additional varied attack data, encompassing known and emerging threats. This dual approach ensures that the system remains relevant and effective against the continuously evolving cyberattacks. The KDD Cup and CICIDS2017 datasets were used to validate this model, which exhibited significant improvements in anomaly detection. It achieved an accuracy of 99.69% on the KDD dataset and 97.93% on the CICIDS2017 dataset, with precision, recall, and F1-scores above 97%, demonstrating its effectiveness in recognizing complex attack patterns. This study contributes significantly to cybersecurity by providing a scalable and adaptable solution for anomaly detection in the face of sophisticated and dynamic cyber threats. The exploration of GANs for data augmentation highlights a promising direction for future research, particularly in situations where data limitations restrict the development of cybersecurity systems. The attention-GAN framework has emerged as a pioneering approach, setting a new benchmark for advanced cyber-defense strategies.

In this dissertation, we propose a systemic framework that prioritizes informative features and examples to enhance each stage of the development process. Specifically, we prioritize informative features and examples and improve the performance of feature learning, data labeling, and data selection. We first propose an approach to extract only informative features that are inherent to solving a target task by using auxiliary out-of-distribution data. We deactivate the noise features in the target distribution by using that in the out-of-distribution data. Next, we introduce an approach that prioritizes informative examples from unlabeled noisy data in order to reduce the labeling cost of active learning. In order to solve the purity-information dilemma, where an attempt to select informative examples induces the selection of many noisy examples, we propose a meta-model that finds the best balance between purity and informativeness. Lastly, we suggest an approach that prioritizes informative examples from labeled noisy data to preserve the performance of data selection. For labeled image noise data, we propose a data selection method that considers the confidence of neighboring samples to maintain the performance of the state-of-the-art Re-labeling models. For labeled text noise data, we present an instruction selection method that takes diversity into account for ranking the quality of instructions with prompting, thereby enhancing the performance of aligned large language models. Overall, our unified framework induces the deep learning development process robust to noisy data, thereby effectively mitigating noisy features and examples in real-world applications.

Geometric Deep Learning techniques have become a transformative force in the field of Computer-Aided Design (CAD), and have the potential to revolutionize how designers and engineers approach and enhance the design process. By harnessing the power of machine learning-based methods, CAD designers can optimize their workflows, save time and effort while making better informed decisions, and create designs that are both innovative and practical. The ability to process the CAD designs represented by geometric data and to analyze their encoded features enables the identification of similarities among diverse CAD models, the proposition of alternative designs and enhancements, and even the generation of novel design alternatives. This survey offers a comprehensive overview of learning-based methods in computer-aided design across various categories, including similarity analysis and retrieval, 2D and 3D CAD model synthesis, and CAD generation from point clouds. Additionally, it provides a complete list of benchmark datasets and their characteristics, along with open-source codes that have propelled research in this domain. The final discussion delves into the challenges prevalent in this field, followed by potential future research directions in this rapidly evolving field.

The classic theory of computation tells us that general computer programs can be converted into equivalent polynomials (albeit over finite fields, which we will focus on implicitly in the sequel). So the pseudo-canonicalization question posed above is equivalent to the pseudo-canonicalization question for general computer programs. Indeed, the question of hiding implementation details within a computer program has a long history, dating at least as far back as the groundbreaking 1976 work of Diffie and Hellman13 introducing the concept of public-key cryptography. Historically, this problem has been called "program obfuscation," albeit it was typically discussed in an ill-defined form. Discussed in these vague terms, it was folklore that truly secure program obfuscation would have revolutionary applications to computing, especially for securing intellectual property.