Since the publication of the first International AI Safety Report, AI capabilities have continued to improve across key domains. New training techniques that teach AI systems to reason step-by-step and inference-time enhancements have primarily driven these advances, rather than simply training larger models. As a result, general-purpose AI systems can solve more complex problems in a range of domains, from scientific research to software development. Their performance on benchmarks that measure performance in coding, mathematics, and answering expert-level science questions has continued to improve, though reliability challenges persist, with systems excelling on some tasks while failing completely on others. These capability improvements also have implications for multiple risks, including risks from biological weapons and cyber attacks. Finally, they pose new challenges for monitoring and controllability. This update examines how AI capabilities have improved since the first Report, then focuses on key risk areas where substantial new evidence warrants updated assessments.

This is the interim publication of the first International Scientific Report on the Safety of Advanced AI. The report synthesises the scientific understanding of general-purpose AI -- AI that can perform a wide variety of tasks -- with a focus on understanding and managing its risks. A diverse group of 75 AI experts contributed to this report, including an international Expert Advisory Panel nominated by 30 countries, the EU, and the UN. Led by the Chair, these independent experts collectively had full discretion over the report's content. The final report is available at arXiv:2501.17805

We introduce a new approach to Topological Data Analysis (TDA) based on Finsler metrics and we also generalize the classical concepts of Vietoris-Rips and Cech complexes within this framework. In particular, we propose a class of directionally dependent Finsler metrics and establish key results demonstrating their relevance to TDA. Moreover, we show that several Information Theoretic perspectives on TDA can often be recovered through the lens of Finsler geometry.

Many continuous wavelets are defined in the frequency domain and do not have analytical expressions in the time domain. Meyer wavelet is ordinarily defined in this way. In this note, we derive new straightforward analytical expressions for both the wavelet and scale function for the Meyer basis. The validity of these expressions is corroborated by numerical computations, yielding no approximation error.

One of the key challenges in many research fields is uncovering how different interconnected systems interact within complex networks, typically represented as multi-layer networks. Capturing the intra- and cross-layer interactions among different domains for analysis and processing calls for topological algebraic descriptors capable of localizing the homologies of different domains, at different scales, according to the learning task. Our first contribution in this paper is to introduce the Cell MultiComplexes (CMCs), which are novel topological spaces that enable the representation of higher-order interactions among interconnected cell complexes. We introduce cross-Laplacian operators as powerful algebraic descriptors of CMC spaces able to capture different topological invariants, whether global or local, at different resolutions. Using the eigenvectors of these operators as bases for the signal representation, we develop topological signal processing tools for signals defined over CMCs. Then, we focus on the signal spectral representation and on the filtering of noisy flows observed over the cross-edges between different layers of CMCs. We show that a local signal representation based on cross-Laplacians yields a better sparsity/accuracy trade-off compared to monocomplex representations, which provide overcomplete representation of local signals. Finally, we illustrate a topology learning strategy designed to infer second-order cross-cells between layers, with applications to brain networks for encoding inter-module connectivity patterns.

In this paper, we study the optimal control for an SEIR model adapted to the vaccination strategy of susceptible individuals. There are factors associated with a vaccination campaign that make this strategy not only a public health issue but also an economic one. In this case, optimal control is important as it minimizes implementation costs. We consider the availability of two vaccines with different efficacy levels, and the control indicates when each vaccine should be used. The optimal strategy specifies in all cases how vaccine purchases should be distributed. For similar efficacy values, we perform a sensitivity analysis on parameters that depend on the intrinsic characteristics of the vaccines. Additionally, we investigate the behavior of the number of infections under the optimal vaccination strategy.

Context: Agility at the business level requires Information Technology (IT) environment flexible and customizable, as well as effective and responsive governance in order to deliver value faster, better, and cheaper to the business. Objective: To understand better this context, our paper seeks to investigate how the domain of agile governance has evolved, as well as to derive implications for research and practice. Method: We conducted a systematic review about the state of art of the agile governance up to and including 2013. Our search strategy identified 1992 studies in 10 databases, of which 167 had the potential to answer our research questions. Results: We organized the studies into four major groups: software engineering, enterprise, manufacturing and multidisciplinary; classifying them into 16 emerging categories. As a result, the review provides a convergent definition for agile governance, six meta- principles, and a map of findings organized by topic and classified by relevance and convergence. Conclusion: The found evidence lead us to believe that agile governance is a relatively new, wide and multidisciplinary area focused on organizational performance and competitiveness that needs to be more intensively studied. Finally, we made improvements and additions to the methodological approach for systematic reviews and qualitative studies.

The regularity of ecosystem size spectra is one of the most intriguing and relevant phenomena on our planet. Aquatic size spectra generally show a log-linearly downtrending shape, following a power-law distribution. A constant log-linear slope has been reported for many marine pelagic ecosystems, often being approximately b = -1. Conversely, there are variable trophic-level-biomass relationships (trophic pyramids). The contrasting observations of a constant biomass (M) spectrum slope and highly variable biomass pyramids may be defined as the constant size spectra-variable trophic dynamics paradox. Here, a 'mass-specific predator-prey-efficiency theory of size spectra' (PETS) is presented and discussed. A thorough analysis of available data, literature and models result in the conclusion that most pelagic marine ecosystems are controlled by trophic processes such as resource-limit stress (bottom-up control) and top-down regulation, with a key role of the maximum carrying capacity of large-sized organisms. This has relevant consequences for the prediction and interpretation of size spectra and the context of fisheries, whaling, and the introduction of exotic predators (e.g., lionfish). The complete size spectrum obtained in situ, including living organisms and non-living particles (e.g., for data from LOPC and UVP) is discussed. This paper is intended as a plea for the integration of modeling approaches, to understand and integrate data and processes across communities including bacteria, phytoplankton, fish and mammals, considering the effect of non-organismic particles.

Structured merge tools exploit programming language syntactic structure to enhance merge accuracy by reducing spurious conflicts reported by unstructured tools. By creating and handling full ASTs, structured tools are language-specific and harder to implement. They can also be computationally expensive when merging large this http URL reduce these drawbacks, semistructured merge tools work with partial ASTs that use strings to represent lower level syntactic structures such as method bodies, and rely on unstructured tools to merge them. This, however, results in merge accuracy loss. To improve accuracy without compromising semistructured merge benefits, we propose a tool that leverages language-specific syntactic separators to infer structure without parsing. We still resort to an unstructured tool to merge lower level structures, but only after preprocessing the code so that text in between separators such as curly braces appear in separate lines. This way we emulate the capabilities of structured merge tools while avoiding their drawbacks. By comparing our tool with a robust implementation of semistructured merge, we find that our tool substantially reduces the number of spurious conflicts. We also observe significant but less substantial reductions on the overall number of reported conflicts, and of files with conflicts. However, similar to structured tools, our tool lets more merge conflicts go undetected. Our tool shows significant improvements over unstructured tools widely used in practice. Finally we observe that exploiting language-specific syntactic separators introduces unique textual alignment challenges.

A new algorithm for voice automatic syllabic splitting in the Portuguese language is proposed, which is based on the envelope of the speech signal of the input audio file. A computational implementation in MatlabTM is presented and made available at the URL this http URL Due to its straightforwardness, the proposed method is very attractive for embedded systems (e.g. i-phones). It can also be used as a screen to assist more sophisticated methods. Voice excerpts containing more than one syllable and identified by the same envelope are named as super-syllables and they are subsequently separated. The results indicate which samples corresponds to the beginning and end of each detected syllable. Preliminary tests were performed to fifty words at an identification rate circa 70% (further improvements may be incorporated to treat particular phonemes). This algorithm is also useful in voice command systems, as a tool in the teaching of Portuguese language or even for patients with speech pathology.

Modeling an ontology is a hard and time-consuming task. Although methodologies are useful for ontologists to create good ontologies, they do not help with the task of evaluating the quality of the ontology to be reused. For these reasons, it is imperative to evaluate the quality of the ontology after constructing it or before reusing it. Few studies usually present only a set of criteria and questions, but no guidelines to evaluate the ontology. The effort to evaluate an ontology is very high as there is a huge dependence on the evaluator's expertise to understand the criteria and questions in depth. Moreover, the evaluation is still very subjective. This study presents a novel methodology for ontology evaluation, taking into account three fundamental principles: i) it is based on the Goal, Question, Metric approach for empirical evaluation; ii) the goals of the methodologies are based on the roles of knowledge representations combined with specific evaluation criteria; iii) each ontology is evaluated according to the type of ontology. The methodology was empirically evaluated using different ontologists and ontologies of the same domain. The main contributions of this study are: i) defining a step-by-step approach to evaluate the quality of an ontology; ii) proposing an evaluation based on the roles of knowledge representations; iii) the explicit difference of the evaluation according to the type of the ontology iii) a questionnaire to evaluate the ontologies; iv) a statistical model that automatically calculates the quality of the ontologies.

Various project management approaches do not consider the impact that uncertainties have on the project. The identified threats by uncertainty in a projec day-to-day are real and immediate and the expectations in a project are often high. The project manager faces a dilemma: decisions must be made in the present about future situations which are inherently uncertain. The use of uncertainty management in project can be a determining factor for the project success. This paper presents a systematic review about uncertainties management in software projects and a guide is proposed based on the review. It aims to present the best practices to manage uncertainties in software projects in a structured way including techniques and strategies to uncertainties containment.

Positionality statements are a reflective practice established in fields such as social sciences, where they enhance transparency, reflexivity, and ethical integrity by acknowledging how researchers identities, experiences, and perspectives may shape their work. This study aimed to investigate the understanding, usage, and potential value of positionality statements in software engineering research, particularly in studies focused on diversity and inclusion.

The motivation of this work is the use of non-invasive and low cost techniques to obtain a faster and more accurate diagnosis of systemic sclerosis (SSc), rheumatic, autoimmune, chronic and rare disease. The technique in question is Near Infrared Spectroscopy (NIRS). Spectra were acquired from three different regions of hand's volunteers. Machine learning algorithms are used to classify and search for the best optical wavelength. The results demonstrate that it is easy to obtain wavelength bands more important for the diagnosis. We use the algorithm RFECV and SVC. The results suggests that the most important wavelength band is at 1270 nm, referring to the luminescence of Singlet Oxygen. The results indicates that the Proximal Interphalangeal Joints region returns better accuracy's scores. Optical spectrometers can be found at low prices and can be easily used in clinical evaluations, while the algorithms used are completely diffused on open source platforms.

Context: Competitiveness is the key to a sustainable development and it demands agility at the business and organizational levels, which in turn requires a flexible and customizable IT environment and effective and responsive governance in order to deliver value to the business. Objective: This paper describes the conceptual development of a theory for analyze and describe agile governance in order to increasing the success rate of their practice, achieving organizational performance and business competitiveness. Method: We adopt a multi-method research, framing the theory conceptual development using Dubin's method of theory building. Results: We have developed a conceptual framework of the theory encompassing its constructs, laws of interaction, boundaries and system states. Conclusion: This theory can provide a better understanding of the nature of agile governance, by mapping of its constructs, mediators, moderators and disturbing factors, in order to help organizations reach better results.

New elliptic cylindrical wavelets are introduced, which exploit the relationship between analysing filters and Floquet's solution of Mathieu differential equations. It is shown that the transfer function of both multiresolution filters is related to the solution of a Mathieu equation of odd characteristic exponent. The number of notches of these analysing filters can be easily designed. Wavelets derived by this method have potential application in the fields of optics, microwaves and electromagnetism.

The use of the open innovation paradigm has been, over the past years, getting special attention in the public sector. Motivated by an urban environment that is increasingly more complex and challenging, several government agencies have been allocating financial resources and efforts to promote open and participative government initiatives. As a way to try and understand this scenario, a systematic review of the literature was conducted, to provide a comprehensive analysis of the scientific papers that were published, seeking to capture, classify, evaluate and synthesize how the use of this paradigm has been put into practice in the public sector. In total, 4,741 preliminary studies were analyzed. From this number, only 37 articles were classified as potentially relevant and moved forward, going through the process of data extraction and analysis. From the data obtained, it was possible to verify that the use of this paradigm started to be reported with a higher frequency in the literature since 2013 and, among the main findings, we highlight the reports of experiences, approach propositions, of understanding how the phenomenon occurs and theoretical reflections. It was also possible to verify that the use of open innovation through social media was one of the pioneer techniques of engagement between the public sector and citizens. In conclusion, the reports confirm that the main challenges of this paradigm applied to the public sector are associated with their respective bureaucratic aspects, therefore lacking a bigger reflection on the procedures and methods to be used in the public sphere.

Although ubiquitous, interactions of groups of individuals (e.g., modern messaging applications, group meetings, or even a parliament discussion) are not yet thoroughly studied. Frequently, single-groups are modeled as critical-mass dynamics, which is a widespread concept used not only by academics but also by politicians and the media. However, less explored questions are how a collection of groups will behave and how the intersection between these groups might change the global dynamics. Here, we formulate this process in terms of binary state dynamics on hypergraphs. We showed that our model has a very rich and unexpected behavior that goes beyond discontinuous transitions. In particular, we might have multistability and intermittency as a consequence of bimodal state distributions. By using artificial random models, we demonstrated that this phenomenology could be associated with community structures. Specifically, we might have multistability or intermittency by controlling the number of bridges between two communities with different densities. The introduction of bridges destroys multistability but creates an intermittent behavior. Furthermore, we provide an analytical formulation showing that the observed pattern for the order parameter and susceptibility are compatible with hybrid phase transitions. Our findings open new paths for research, ranging from physics, on the formal calculation of quantities of interest, to social sciences, where new experiments can be designed.

A new interpretation for the wavelet analysis is reported, which can is viewed as an information processing technique. It was recently proposed that every basic wavelet could be associated with a proper probability density, allowing defining the entropy of a wavelet. Introducing now the concept of wavelet mutual information between a signal and an analysing wavelet fulfils the foundations of a wavelet information theory (WIT). Both continuous and discrete time signals are considered. Finally, we showed how to compute the information provided by a multiresolution analysis by means of the inhomogeneous wavelet expansion. Highlighting ideas behind the WIT are presented.