Directly learning from examples of random difficulty levels is often challenging for both humans and machine learning models. A more effective strategy involves exposing learners to examples in a progressive order, from easy to difficult. Curriculum Learning (CL) has been proposed to implement this strategy in machine learning model training. However, two key challenges persist in CL framework design: defining the difficulty of training data and determining the appropriate amount of data to input at each training step. This paper presents a Psychology-based Unified Dynamic Framework for Curriculum Learning (PUDF), drawing inspiration from psychometrics. We quantify the difficulty of training data by applying Item Response Theory (IRT) to responses from Artificial Crowds (AC). This theory-driven IRT-AC approach leads to global (i.e., model-independent) and interpretable difficulty values. Leveraging IRT, we propose a Dynamic Data Selection via Model Ability Estimation (DDS-MAE) strategy to schedule the appropriate amount of data during model training. Since our difficulty labeling and model ability estimation are based on a consistent theory, namely IRT, their values are comparable within the same scope, potentially leading to a faster convergence compared to the other CL methods. Experimental results demonstrate that fine-tuning pre-trained language models with PUDF enhances their performance on the GLUE benchmark. Moreover, PUDF surpasses other state-of-the-art (SOTA) CL methods on the GLUE benchmark. We further explore the components of PUDF, namely the difficulty measurer (IRT-AC) and the training scheduler (DDS-MAE) qualitatively and quantitatively. Lastly, we conduct an ablation study to clarify which components of PUDF contribute to faster convergence and higher accuracy.

The Exame Nacional do Ensino M\'edio (ENEM) is a pivotal test for Brazilian students, required for admission to a significant number of universities in Brazil. The test consists of four objective high-school level tests on Math, Humanities, Natural Sciences and Languages, and one writing essay. Students' answers to the test and to the accompanying socioeconomic status questionnaire are made public every year (albeit anonymized) due to transparency policies from the Brazilian Government. In the context of large language models (LLMs), these data lend themselves nicely to comparing different groups of humans with AI, as we can have access to human and machine answer distributions. We leverage these characteristics of the ENEM dataset and compare GPT-3.5 and 4, and MariTalk, a model trained using Portuguese data, to humans, aiming to ascertain how their answers relate to real societal groups and what that may reveal about the model biases. We divide the human groups by using socioeconomic status (SES), and compare their answer distribution with LLMs for each question and for the essay. We find no significant biases when comparing LLM performance to humans on the multiple-choice Brazilian Portuguese tests, as the distance between model and human answers is mostly determined by the human accuracy. A similar conclusion is found by looking at the generated text as, when analyzing the essays, we observe that human and LLM essays differ in a few key factors, one being the choice of words where model essays were easily separable from human ones. The texts also differ syntactically, with LLM generated essays exhibiting, on average, smaller sentences and less thought units, among other differences. These results suggest that, for Brazilian Portuguese in the ENEM context, LLM outputs represent no group of humans, being significantly different from the answers from Brazilian students across all tests.

In this paper, we conduct an empirical analysis of how large language models (LLMs), specifically GPT-4, interpret constitutional principles in complex decision-making scenarios. We examine rulings from the Italian Constitutional Court on bioethics issues that involve trade-offs between competing values and compare model-generated legal arguments on these issues to those presented by the State, the Court, and the applicants. Our results indicate that GPT-4 consistently aligns more closely with progressive interpretations of the Constitution, often overlooking competing values and mirroring the applicants' views rather than the more conservative perspectives of the State or the Court's moderate positions. Our experiments reveal a distinct tendency of GPT-4 to favor progressive legal interpretations, underscoring the influence of underlying data biases. We thus underscore the importance of testing alignment in real-world scenarios and considering the implications of deploying LLMs in decision-making processes.

The electric vehicle (EV) battery supply chain's vulnerability to disruptions necessitates advanced predictive analytics. We present SHIELD (Schema-based Hierarchical Induction for EV supply chain Disruption), a system integrating Large Language Models (LLMs) with domain expertise for EV battery supply chain risk assessment. SHIELD combines: (1) LLM-driven schema learning to construct a comprehensive knowledge library, (2) a disruption analysis system utilizing fine-tuned language models for event extraction, multi-dimensional similarity matching for schema matching, and Graph Convolutional Networks (GCNs) with logical constraints for prediction, and (3) an interactive interface for visualizing results and incorporating expert feedback to enhance decision-making. Evaluated on 12,070 paragraphs from 365 sources (2022-2023), SHIELD outperforms baseline GCNs and LLM+prompt methods (e.g., GPT-4o) in disruption prediction. These results demonstrate SHIELD's effectiveness in combining LLM capabilities with domain expertise for enhanced supply chain risk assessment.

Predictions often influence the reality which they aim to predict, an effect known as performativity. Existing work focuses on accuracy maximization under this effect, but model deployment may have important unintended impacts, especially in multiagent scenarios. In this work, we investigate performative prediction in a concrete game-theoretic setting where social welfare is an alternative objective to accuracy maximization. We explore a collective risk dilemma scenario where maximising accuracy can negatively impact social welfare, when predicting collective behaviours. By assuming knowledge of a Bayesian agent behavior model, we then show how to achieve better trade-offs and use them for mechanism design.

It is generally accepted that all cyber attacks cannot be prevented, creating a need for the ability to detect and respond to cyber attacks. Both connectionist and symbolic AI are currently being used to support such detection and response. In this paper, we make the case for combining them using neurosymbolic AI. We identify a set of challenges when using AI today and propose a set of neurosymbolic use cases we believe are both interesting research directions for the neurosymbolic AI community and can have an impact on the cyber security field. We demonstrate feasibility through two proof-of-concept experiments.

Cyber-attack attribution is an important process that allows experts to put in place attacker-oriented countermeasures and legal actions. The analysts mainly perform attribution manually, given the complex nature of this task. AI and, more specifically, Natural Language Processing (NLP) techniques can be leveraged to support cybersecurity analysts during the attribution process. However powerful these techniques are, they need to deal with the lack of datasets in the attack attribution domain. In this work, we will fill this gap and will provide, to the best of our knowledge, the first dataset on cyber-attack attribution. We designed our dataset with the primary goal of extracting attack attribution information from cybersecurity texts, utilizing named entity recognition (NER) methodologies from the field of NLP. Unlike other cybersecurity NER datasets, ours offers a rich set of annotations with contextual details, including some that span phrases and sentences. We conducted extensive experiments and applied NLP techniques to demonstrate the dataset's effectiveness for attack attribution. These experiments highlight the potential of Large Language Models (LLMs) capabilities to improve the NER tasks in cybersecurity datasets for cyber-attack attribution.

In-context learning (ICL) greatly improves the performance of large language models (LLMs) on various down-stream tasks, where the improvement highly depends on the quality of demonstrations. In this work, we introduce syntactic knowledge to select better in-context examples for machine translation (MT). We propose a new strategy, namely Syntax-augmented COverage-based In-context example selection (SCOI), leveraging the deep syntactic structure beyond conventional word matching. Specifically, we measure the set-level syntactic coverage by computing the coverage of polynomial terms with the help of a simplified tree-to-polynomial algorithm, and lexical coverage using word overlap. Furthermore, we devise an alternate selection approach to combine both coverage measures, taking advantage of syntactic and lexical information. We conduct experiments with two multi-lingual LLMs on six translation directions. Empirical results show that our proposed SCOI obtains the highest average COMET score among all learning-free methods, indicating that combining syntactic and lexical coverage successfully helps to select better in-context examples for MT.

Elon Musk's SpaceX and its competitors are making reliable, and decently-fast satellite internet services a reality thanks to a growing armada of shimmering satellites orbiting overhead. Through its constellation of over 6,000, 500-pound satellites, SpaceX's Starlink internet service already reportedly provides broadband to around three million global users, some in remote locations underserved by traditional internet providers. But what happens when all those aging satellites no longer serve their purpose? A new report from environmentally-focused advocacy group PIRG warns the current approach to decommissioning old satellites, which usually involves having them burn to a crisp when re-entering the atmosphere, lacks meaningful rules and regulation. That absence of oversight, they say, could lead to an increase in dangerous space junk affecting Earth, especially as competing satellite internet companies rush to build out and launch tens of thousands of new satellites into orbit.

In recent years, the threat facing airports from growing and increasingly sophisticated cyberattacks has become evident. Airports are considered a strategic national asset, so protecting them from attacks, specifically cyberattacks, is a crucial mission. One way to increase airports' security is by using Digital Twins (DTs). This paper shows and demonstrates how DTs can enhance the security mission. The integration of DTs with Generative AI (GenAI) algorithms can lead to synergy and new frontiers in fighting cyberattacks. The paper exemplifies ways to model cyberattack scenarios using simulations and generate synthetic data for testing defenses. It also discusses how DTs can be used as a crucial tool for vulnerability assessment by identifying weaknesses, prioritizing, and accelerating remediations in case of cyberattacks. Moreover, the paper demonstrates approaches for anomaly detection and threat hunting using Machine Learning (ML) and GenAI algorithms. Additionally, the paper provides impact prediction and recovery coordination methods that can be used by DT operators and stakeholders. It also introduces ways to harness the human factor by integrating training and simulation algorithms with Explainable AI (XAI) into the DT platforms. Lastly, the paper offers future applications and technologies that can be utilized in DT environments.