Search

Chord progressions encapsulate important information about music, pertaining to its structure and conveyed emotions. They serve as the backbone of musical composition, and in many cases, they are the sole information required for a musician to play along and follow the music. Despite their importance, chord progressions as a data domain remain underexplored. There is a lack of large-scale datasets suitable for deep learning applications, and limited research exploring chord progressions as an input modality. In this work, we present Chordonomicon, a dataset of over 666,000 songs and their chord progressions, annotated with structural parts, genre, and release date - created by scraping various sources of user-generated progressions and associated metadata. We demonstrate the practical utility of the Chordonomicon dataset for classification and generation tasks, and discuss its potential to provide valuable insights to the research community. Chord progressions are unique in their ability to be represented in multiple formats (e.g. text, graph) and the wealth of information chords convey in given contexts, such as their harmonic function . These characteristics make the Chordonomicon an ideal testbed for exploring advanced machine learning techniques, including transformers, graph machine learning, and hybrid systems that combine knowledge representation and machine learning.

Large language models have evolved to process multiple modalities beyond text, such as images and audio, which motivates us to explore how to effectively leverage them for graph machine learning tasks. The key question, therefore, is how to transform graphs into linear sequences of tokens, a process we term graph linearization, so that LLMs can handle graphs naturally. We consider that graphs should be linearized meaningfully to reflect certain properties of natural language text, such as local dependency and global alignment, in order to ease contemporary LLMs, trained on trillions of textual tokens, better understand graphs. To achieve this, we developed several graph linearization methods based on graph centrality, degeneracy, and node relabeling schemes. We then investigated their effect on LLM performance in graph reasoning tasks. Experimental results on synthetic graphs demonstrate the effectiveness of our methods compared to random linearization baselines. Our work introduces novel graph representations suitable for LLMs, contributing to the potential integration of graph machine learning with the trend of multi-modal processing using a unified transformer model.

We describe our contribution to the Strict and Strict-Small tracks of the 2nd iteration of the BabyLM Challenge. The shared task is centered around efficient pre-training given data constraints motivated by human development. In response, we study the effect of synthetic story data in language pre-training using TinyStories: a recently introduced dataset of short stories. Initially, we train GPT-Neo models on subsets of TinyStories, while varying the amount of available data. We find that, even with access to less than 100M words, the models are able to generate high-quality, original completions to a given story, and acquire substantial linguistic knowledge. To measure the effect of synthetic story data, we train LTG-BERT encoder models on a combined dataset of: a subset of TinyStories, story completions generated by GPT-Neo, and a subset of the BabyLM dataset. Our experimentation reveals that synthetic data can occasionally offer modest gains, but overall have a negative influence on linguistic understanding. Our work offers an initial study on synthesizing story data in low resource settings and underscores their potential for augmentation in data-constrained language modeling. We publicly release our models and implementation on our GitHub.

Riddle-solving requires advanced reasoning skills, pushing LLMs to engage in abstract thinking and creative problem-solving, often revealing limitations in their cognitive abilities. In this paper, we examine the riddle-solving capabilities of LLMs using a multiple-choice format, exploring how different prompting techniques impact performance on riddles that demand diverse reasoning skills. To enhance results, we introduce RISCORE (RIddle Solving with COntext REcontruciton) a novel fully automated prompting method that generates and utilizes contextually reconstructed sentence-based puzzles in conjunction with the original examples to create few-shot exemplars. Our experiments demonstrate that RISCORE significantly improves the performance of language models in both vertical and lateral thinking tasks, surpassing traditional exemplar selection strategies across a variety of few-shot settings., Comment: Accepted at COLING 2025

Equivocation and ambiguity in public speech are well-studied discourse phenomena, especially in political science and analysis of political interviews. Inspired by the well-grounded theory on equivocation, we aim to resolve the closely related problem of response clarity in questions extracted from political interviews, leveraging the capabilities of Large Language Models (LLMs) and human expertise. To this end, we introduce a novel taxonomy that frames the task of detecting and classifying response clarity and a corresponding clarity classification dataset which consists of question-answer (QA) pairs drawn from political interviews and annotated accordingly. Our proposed two-level taxonomy addresses the clarity of a response in terms of the information provided for a given question (high-level) and also provides a fine-grained taxonomy of evasion techniques that relate to unclear, ambiguous responses (lower-level). We combine ChatGPT and human annotators to collect, validate and annotate discrete QA pairs from political interviews, to be used for our newly introduced response clarity task. We provide a detailed analysis and conduct several experiments with different model architectures, sizes and adaptation methods to gain insights and establish new baselines over the proposed dataset and task., Comment: Accepted at Findings of EMNLP 2024

Gender bias in machine translation (MT) systems poses significant challenges that often result in the reinforcement of harmful stereotypes. Especially in the labour domain where frequently occupations are inaccurately associated with specific genders, such biases perpetuate traditional gender stereotypes with a significant impact on society. Addressing these issues is crucial for ensuring equitable and accurate MT systems. This paper introduces a novel approach to studying occupation-related gender bias through the creation of the GOSt-MT (Gender and Occupation Statistics for Machine Translation) Knowledge Graph. GOSt-MT integrates comprehensive gender statistics from real-world labour data and textual corpora used in MT training. This Knowledge Graph allows for a detailed analysis of gender bias across English, French, and Greek, facilitating the identification of persistent stereotypes and areas requiring intervention. By providing a structured framework for understanding how occupations are gendered in both labour markets and MT systems, GOSt-MT contributes to efforts aimed at making MT systems more equitable and reducing gender biases in automated translations., Comment: Accepted at the KG-STAR'24: Workshop on Knowledge Graphs for Responsible AI co-located with the 33rd ACM CIKM Conference, October 25, 2024, Boise, Idaho

Multimodal models are critical for music understanding tasks, as they capture the complex interplay between audio and lyrics. However, as these models become more prevalent, the need for explainability grows-understanding how these systems make decisions is vital for ensuring fairness, reducing bias, and fostering trust. In this paper, we introduce MusicLIME, a model-agnostic feature importance explanation method designed for multimodal music models. Unlike traditional unimodal methods, which analyze each modality separately without considering the interaction between them, often leading to incomplete or misleading explanations, MusicLIME reveals how audio and lyrical features interact and contribute to predictions, providing a holistic view of the model's decision-making. Additionally, we enhance local explanations by aggregating them into global explanations, giving users a broader perspective of model behavior. Through this work, we contribute to improving the interpretability of multimodal music models, empowering users to make informed choices, and fostering more equitable, fair, and transparent music understanding systems., Comment: GitHub repository: https://github.com/IamTheo2000/MusicLIME

The surge of state-of-the-art Transformer-based models has undoubtedly pushed the limits of NLP model performance, excelling in a variety of tasks. We cast the spotlight on the underexplored task of Natural Language Inference (NLI), since models trained on popular well-suited datasets are susceptible to adversarial attacks, allowing subtle input interventions to mislead the model. In this work, we validate the usage of natural language explanation as a model-agnostic defence strategy through extensive experimentation: only by fine-tuning a classifier on the explanation rather than premise-hypothesis inputs, robustness under various adversarial attacks is achieved in comparison to explanation-free baselines. Moreover, since there is no standard strategy of testing the semantic validity of the generated explanations, we research the correlation of widely used language generation metrics with human perception, in order for them to serve as a proxy towards robust NLI models. Our approach is resource-efficient and reproducible without significant computational limitations.

Despite the rapid evolution and increasing efficacy of language and vision generative models, there remains a lack of comprehensive datasets that bridge the gap between personalized fashion needs and AI-driven design, limiting the potential for truly inclusive and customized fashion solutions. In this work, we leverage generative models to automatically construct a fashion image dataset tailored to various occasions, styles, and body types as instructed by users. We use different Large Language Models (LLMs) and prompting strategies to offer personalized outfits of high aesthetic quality, detail, and relevance to both expert and non-expert users' requirements, as demonstrated by qualitative analysis. Up until now the evaluation of the generated outfits has been conducted by non-expert human subjects. Despite the provided fine-grained insights on the quality and relevance of generation, we extend the discussion on the importance of expert knowledge for the evaluation of artistic AI-generated datasets such as this one. Our dataset is publicly available on GitHub at https://github.com/georgiarg/Prompt2Fashion.

In the rapidly evolving fields of natural language processing and computer vision, Visual Word Sense Disambiguation (VWSD) stands as a critical, yet challenging task. The quest for models that can seamlessly integrate and interpret multimodal data is more pressing than ever. Imagine a system that can understand language with the depth and nuance of human cognition, while simultaneously interpreting the rich visual context of the world around it. We present ARPA, an architecture that fuses the unparalleled contextual understanding of large language models with the advanced feature extraction capabilities of transformers, which then pass through a custom Graph Neural Network (GNN) layer to learn intricate relationships and subtle nuances within the data. This innovative architecture not only sets a new benchmark in visual word disambiguation but also introduces a versatile framework poised to transform how linguistic and visual data interact by harnessing the synergistic strengths of its components, ensuring robust performance even in the most complex disambiguation scenarios. Through a series of experiments and comparative analysis, we reveal the substantial advantages of our model, underscoring its potential to redefine standards in the field. Beyond its architectural prowess, our architecture excels through experimental enrichments, including sophisticated data augmentation and multi-modal training techniques. ARPA's introduction marks a significant milestone in visual word disambiguation, offering a compelling solution that bridges the gap between linguistic and visual modalities. We invite researchers and practitioners to explore the capabilities of our model, envisioning a future where such hybrid models drive unprecedented advancements in artificial intelligence.

As NLP models become increasingly integral to decision-making processes, the need for explainability and interpretability has become paramount. In this work, we propose a framework that achieves the aforementioned by generating semantically edited inputs, known as counterfactual interventions, which change the model prediction, thus providing a form of counterfactual explanations for the model. We test our framework on two NLP tasks - binary sentiment classification and topic classification - and show that the generated edits are contrastive, fluent and minimal, while the whole process remains significantly faster that other state-of-the-art counterfactual editors.

The advent of artificial intelligence has contributed in a groundbreaking transformation of the fashion industry, redefining creativity and innovation in unprecedented ways. This work investigates methodologies for generating tailored fashion descriptions using two distinct Large Language Models and a Stable Diffusion model for fashion image creation. Emphasizing adaptability in AI-driven fashion creativity, we depart from traditional approaches and focus on prompting techniques, such as zero-shot and few-shot learning, as well as Chain-of-Thought (CoT), which results in a variety of colors and textures, enhancing the diversity of the outputs. Central to our methodology is Retrieval-Augmented Generation (RAG), enriching models with insights from fashion sources to ensure contemporary representations. Evaluation combines quantitative metrics such as CLIPscore with qualitative human judgment, highlighting strengths in creativity, coherence, and aesthetic appeal across diverse styles. Among the participants, RAG and few-shot learning techniques are preferred for their ability to produce more relevant and appealing fashion descriptions. Our code is provided at https://github.com/georgiarg/AutoFashion.

As machine learning (ML) models and datasets increase in complexity, the demand for methods that enhance explainability and interpretability becomes paramount. Prototypes, by encapsulating essential characteristics within data, offer insights that enable tactical decision-making and enhance transparency. Traditional prototype methods often rely on sub-symbolic raw data and opaque latent spaces, reducing explainability and increasing the risk of misinterpretations. This paper presents a novel framework that utilizes semantic descriptions to define prototypes and provide clear explanations, effectively addressing the shortcomings of conventional methods. Our approach leverages concept-based descriptions to cluster data on the semantic level, ensuring that prototypes not only represent underlying properties intuitively but are also straightforward to interpret. Our method simplifies the interpretative process and effectively bridges the gap between complex data structures and human cognitive processes, thereby enhancing transparency and fostering trust. Our approach outperforms existing widely-used prototype methods in facilitating human understanding and informativeness, as validated through a user survey., Comment: This paper has been accepted for publication as a full paper at the 33rd ACM International Conference on Information and Knowledge Management (CIKM 2024)

For nearly a century, social researchers and psychologists have debated the efficacy of psychometric scales for attitude measurement, focusing on Thurstone's equal appearing interval scales and Likert's summated rating scales. Thurstone scales fell out of favour due to the labour intensive process of gathering judges' opinions on the initial items. However, advancements in technology have mitigated these challenges, nullifying the simplicity advantage of Likert scales, which have their own methodological issues. This study explores a methodological experiment to develop a Thurstone scale for assessing attitudes towards individuals living with AIDS. An electronic questionnaire was distributed to a group of judges, including undergraduate, postgraduate, and PhD students from disciplines such as social policy, law, medicine, and computer engineering, alongside established social researchers, and their responses were statistically analysed. The primary innovation of this study is the incorporation of an Artificial Intelligence (AI) Large Language Model (LLM) to evaluate the initial 63 items, comparing its assessments with those of the human judges. Interestingly, the AI provided also detailed explanations for its categorisation. Results showed no significant difference between AI and human judges for 35 items, minor differences for 23 items, and major differences for 5 items. This experiment demonstrates the potential of integrating AI with traditional psychometric methods to enhance the development of attitude measurement scales.

The massive digitization of artworks during the last decades created the need for categorization, analysis, and management of huge amounts of data related to abstract concepts, highlighting a challenging problem in the field of computer science. The rapid progress of artificial intelligence and neural networks has provided tools and technologies that seem worthy of the challenge. Recognition of various art features in artworks has gained attention in the deep learning society. In this paper, we are concerned with the problem of art style recognition using deep networks. We compare the performance of 8 different deep architectures (VGG16, VGG19, ResNet50, ResNet152, Inception-V3, DenseNet121, DenseNet201 and Inception-ResNet-V2), on two different art datasets, including 3 architectures that have never been used on this task before, leading to state-of-the-art performance. We study the effect of data preprocessing prior to applying a deep learning model. We introduce a stacking ensemble method combining the results of first-stage classifiers through a meta-classifier, with the innovation of a versatile approach based on multiple models that extract and recognize different characteristics of the input, creating a more consistent model compared to existing works and achieving state-of-the-art accuracy on the largest art dataset available (WikiArt - 68,55%). We also discuss the impact of the data and art styles themselves on the performance of our models forming a manifold perspective on the problem.

Using the measured and projected invisible mass spectrum of the $K^+\to\pi^+\nu\bar\nu$ mode, we determine the current and future constraints within the model-independent framework of the weak effective theory at dimension-six. We work in two different operator bases depending whether neutrinos are Majorana or Dirac fermions. This makes it possible to transparently incorporate mass effects of additional sterile neutrinos for all operators., Comment: Prepared for the 2024 Electroweak session of the 58th Rencontres de Moriond, 5 pages, 4 figures

Light dark matter with flavor-violating couplings to fermions may be copiously produced in the laboratory as missing energy from decays of SM particles. Here we study the effective Lagrangian of a light dark vector with generic dipole or vector couplings. We calculate the resulting two-body decay rates of mesons, baryons and leptons as a function of the dark vector mass and show that existing experimental limits probe UV scales as large as $10^{12} \,\mathrm{GeV}$. We also derive the general RGEs in order to constrain the flavor-universal UV scenario, where all flavor violation arises radiatively proportional to the CKM matrix., Comment: 40 pages, 11 figures, 6 tables

In this study, we present an analysis of the fine-tuning required in various inflationary models in order to explain the production of Primordial Black Holes (PBHs). We specifically examine the degree of fine-tuning necessary in two prominent single field inflationary models: those with an inflection point and those with step-like features in the potential. Our findings indicate that models with step-like features generally require less fine-tuning compared to those with an inflection point, making them more viable for consistent PBH production. An interesting outcome of these models is that, in addition to improved fine-tuning, they may also predict low-frequency signals that can be detected by pulsar timing array (PTA) collaborations. Additionally, we extend our analysis to multifield inflationary models to assess whether the integration of additional fields can further alleviate the fine-tuning demands. The study also explores the role of a spectator field and its impact on the fine-tuning process. Through a comparative analysis across these models, we evaluate their parametric sensitivities and alignment with the constraints from Cosmic Microwave Background (CMB) observations. Our results indicate that although mechanisms involving a spectator field can circumvent the issue of fine-tuning parameters for PBH production, both multifield models and models with step-like features present promising alternatives. While fine-tuning involves multiple considerations, our primary objective is to evaluate various inflationary models to identify the one that most naturally explains the formation of PBHs. Hence, this study aims to the development of less constrained inflationary scenarios that align with observational data and support the theoretical possibility of PBH production., Comment: 22 pages, 8 figures

Explainable Artificial Intelligence (XAI) has emerged as a critical area of research aimed at enhancing the transparency and interpretability of AI systems. Counterfactual Explanations (CFEs) offer valuable insights into the decision-making processes of machine learning algorithms by exploring alternative scenarios where certain factors differ. Despite the growing popularity of CFEs in the XAI community, existing literature often overlooks the diverse needs and objectives of users across different applications and domains, leading to a lack of tailored explanations that adequately address the different use cases. In this paper, we advocate for a nuanced understanding of CFEs, recognizing the variability in desired properties based on user objectives and target applications. We identify three primary user objectives and explore the desired characteristics of CFEs in each case. By addressing these differences, we aim to design more effective and tailored explanations that meet the specific needs of users, thereby enhancing collaboration with AI systems.

A novel methodology for analysing the relation between the energy density in gravitational waves and primordial power spectra is developed. Focusing on scalar-induced gravitational radiation, this methodology is applied to a number of scenarios for the primordial black hole formation. Being differed from conventional Bayesian approaches, its advantages include directness and computational efficiency, which are crucial for handling the complex data characteristic of gravitational wave research. As an important application, it is demonstrated that it allows to systematically identify all scenarios consistent with current and future pulsar-timing-array data., Comment: 12 pages, 7 figures

We investigate the large curvature perturbations which can lead to the formation of primordial black holes (PBHs) in the context of no-scale supergravity. Our study does not depend on any exotic scenario, such as scalar potentials with inflection points or bulks, and aims to avoid the fine-tuning of model parameters to achieve the formation of PBHs. This formation relies on the quantum fluctuations of a light spectator stochastic field after the inflationary period. Our analysis is based on the SU(2,1)/SU(2)$\times$U(1) symmetry, considering both the inflaton and the spectator field. Specifically, we examine existing no-scale models with Starobinsky-like scalar potentials that are consistent with observable constraints on inflation from measurements of the cosmic microwave background (CMB). These models involve two chiral fields: the inflaton and the modulus field. We propose a novel role for the modulus field as a spectator field, responsible for generating PBHs. Our hypothesis suggests that while the inflaton field satisfies the CMB constraints of inflation, it is the modulus field acting as the spectator that leads to large curvature perturbations, capable to explain the production of PBHs. Additionally, we prioritize retaining the inflationary constraints from the CMB through the consideration of spectator fluctuations. Therefore, by exploring the relationship between these fields within the framework of the SU(2,1)/SU(2)$\times$U(1) symmetry, our aim is to unveil their implications for the formation of PBHs., Comment: 12 pages, 10 figures

In this paper, we outline our submission for the SemEval-2024 Task 9 competition: 'BRAINTEASER: A Novel Task Defying Common Sense'. We engage in both sub-tasks: Sub-task A-Sentence Puzzle and Sub-task B-Word Puzzle. We evaluate a plethora of pre-trained transformer-based language models of different sizes through fine-tuning. Subsequently, we undertake an analysis of their scores and responses to aid future researchers in understanding and utilizing these models effectively. Our top-performing approaches secured competitive positions on the competition leaderboard across both sub-tasks. In the evaluation phase, our best submission attained an average accuracy score of 81.7% in the Sentence Puzzle, and 85.4% in the Word Puzzle, significantly outperforming the best neural baseline (ChatGPT) by more than 20% and 30% respectively., Comment: SemEval-2024

In this paper, we present our team's submissions for SemEval-2024 Task-6 - SHROOM, a Shared-task on Hallucinations and Related Observable Overgeneration Mistakes. The participants were asked to perform binary classification to identify cases of fluent overgeneration hallucinations. Our experimentation included fine-tuning a pre-trained model on hallucination detection and a Natural Language Inference (NLI) model. The most successful strategy involved creating an ensemble of these models, resulting in accuracy rates of 77.8% and 79.9% on model-agnostic and model-aware datasets respectively, outperforming the organizers' baseline and achieving notable results when contrasted with the top-performing results in the competition, which reported accuracies of 84.7% and 81.3% correspondingly., Comment: SemEval-2024

Story Visualization (SV) is a challenging generative vision task, that requires both visual quality and consistency between different frames in generated image sequences. Previous approaches either employ some kind of memory mechanism to maintain context throughout an auto-regressive generation of the image sequence, or model the generation of the characters and their background separately, to improve the rendering of characters. On the contrary, we embrace a completely parallel transformer-based approach, exclusively relying on Cross-Attention with past and future captions to achieve consistency. Additionally, we propose a Character Guidance technique to focus on the generation of characters in an implicit manner, by forming a combination of text-conditional and character-conditional logits in the logit space. We also employ a caption-augmentation technique, carried out by a Large Language Model (LLM), to enhance the robustness of our approach. The combination of these methods culminates into state-of-the-art (SOTA) results over various metrics in the most prominent SV benchmark (Pororo-SV), attained with constraint resources while achieving superior computational complexity compared to previous arts. The validity of our quantitative results is supported by a human survey.

Counterfactual explanations (CEs) based on concepts are explanations that consider alternative scenarios to understand which high-level semantic features contributed to particular model predictions. In this work, we propose CEs based on the semantic graphs accompanying input data to achieve more descriptive, accurate, and human-aligned explanations. Building upon state-of-the-art (SoTA) conceptual attempts, we adopt a model-agnostic edit-based approach and introduce leveraging GNNs for efficient Graph Edit Distance (GED) computation. With a focus on the visual domain, we represent images as scene graphs and obtain their GNN embeddings to bypass solving the NP-hard graph similarity problem for all input pairs, an integral part of the CE computation process. We apply our method to benchmark and real-world datasets with varying difficulty and availability of semantic annotations. Testing on diverse classifiers, we find that our CEs outperform previous SoTA explanation models based on semantics, including both white and black-box as well as conceptual and pixel-level approaches. Their superiority is proven quantitatively and qualitatively, as validated by human subjects, highlighting the significance of leveraging semantic edges in the presence of intricate relationships. Our model-agnostic graph-based approach is widely applicable and easily extensible, producing actionable explanations across different contexts.

Exploring the capabilities of Large Language Models (LLMs) in puzzle solving unveils critical insights into their potential and challenges in AI, marking a significant step towards understanding their applicability in complex reasoning tasks. This survey leverages a unique taxonomy -- dividing puzzles into rule-based and rule-less categories -- to critically assess LLMs through various methodologies, including prompting techniques, neuro-symbolic approaches, and fine-tuning. Through a critical review of relevant datasets and benchmarks, we assess LLMs' performance, identifying significant challenges in complex puzzle scenarios. Our findings highlight the disparity between LLM capabilities and human-like reasoning, particularly in those requiring advanced logical inference. The survey underscores the necessity for novel strategies and richer datasets to advance LLMs' puzzle-solving proficiency and contribute to AI's logical reasoning and creative problem-solving advancements.

Counterfactuals have been established as a popular explainability technique which leverages a set of minimal edits to alter the prediction of a classifier. When considering conceptual counterfactuals on images, the edits requested should correspond to salient concepts present in the input data. At the same time, conceptual distances are defined by knowledge graphs, ensuring the optimality of conceptual edits. In this work, we extend previous endeavors on graph edits as counterfactual explanations by conducting a comparative study which encompasses both supervised and unsupervised Graph Neural Network (GNN) approaches. To this end, we pose the following significant research question: should we represent input data as graphs, which is the optimal GNN approach in terms of performance and time efficiency to generate minimal and meaningful counterfactual explanations for black-box image classifiers?

We present MaRTIn, an extendable all-in-one package for calculating amplitudes up to two loops in an expansion in external momenta or using the method of infrared rearrangement. Renormalizable and non-renormalizable models can be supplied by the user; an implementation of the Standard Model is included in the package. In this manual, we discuss the scope and functionality of the software, and give instructions of its use., Comment: 35 pages, 1 figure, source code available under https://gitlab.com/manstam/martin

In the age of music streaming platforms, the task of automatically tagging music audio has garnered significant attention, driving researchers to devise methods aimed at enhancing performance metrics on standard datasets. Most recent approaches rely on deep neural networks, which, despite their impressive performance, possess opacity, making it challenging to elucidate their output for a given input. While the issue of interpretability has been emphasized in other fields like medicine, it has not received attention in music-related tasks. In this study, we explored the relevance of interpretability in the context of automatic music tagging. We constructed a workflow that incorporates three different information extraction techniques: a) leveraging symbolic knowledge, b) utilizing auxiliary deep neural networks, and c) employing signal processing to extract perceptual features from audio files. These features were subsequently used to train an interpretable machine-learning model for tag prediction. We conducted experiments on two datasets, namely the MTG-Jamendo dataset and the GTZAN dataset. Our method surpassed the performance of baseline models in both tasks and, in certain instances, demonstrated competitiveness with the current state-of-the-art. We conclude that there are use cases where the deterioration in performance is outweighed by the value of interpretability., Comment: Github Repository: https://github.com/vaslyb/perceptible-music-tagging

We investigate the viability of primordial black hole (PBH) formation in the Standard Model (SM) in a scenario that does not rely on specific inflationary features or any exotic physics such as phase transitions or non-minimal coupling to gravity. If the Brout-Englert-Higgs (BEH) field lies exactly at the transition between metastability and stability, its potential exhibits an inflexion point due to radiative corrections. The BEH can act like a stochastic curvaton field, leading to a non-Gaussian tail of large curvature fluctuations that later collapse into PBHs when they re-enter inside the horizon. This scenario would require a precise value of the top-quark mas to ensure the Higgs stability, which is disfavored but still consistent with the most recent measurements. However, we also find that large curvature fluctuations are also generated on cosmological scales that are inconsistent with cosmic microwave background (CMB) observations. We therefore conclude that the SM cannot have led to the formation of PBHs based on this mechanism. Nevertheless, a variation of the scenario based on the Palatini formulation of gravity may have provided the conditions to produce stellar-mass PBHs with an abundance comparable to dark matter, without producing too large curvature fluctuations on cosmological scales., Comment: 7 pages, 7 figures

The excellent experimental prospects to measure the invisible mass spectrum of the $K^+\to\pi^+\nu\bar\nu$ decay opens a new path to test generalised quark--neutrino interactions with flavour changing $s\to d$ transitions and as such to novel probes of Physics beyond the Standard Model. Such signals can be a consequence of new lepton-number violating or lepton-number conserving interactions, with their interpretations depending on the Majorana versus Dirac nature of the neutrinos. Furthermore, the possible existence of new massive sterile neutrinos can be tested via their distinctive imprints in the invariant mass spectrum. Within the model-independent framework of the weak effective theory at dimension-six, we study the New Physics effects of Majorana and Dirac neutrinos on the differential distribution of $K^+\rightarrow \pi^+\nu\bar\nu$ allowing for lepton-number violating interactions and potential new sterile neutrinos. We determine the current and expected future sensitivity on the corresponding $\Delta S=1$ neutral-current Wilson coefficients using the distribution measured by the NA62 collaboration and accounting for expected improvements based on the HIKE experiment. We present single-operator fits and also determine correlations among different type of operators. Even though we focus on $s\to d\nu\nu$ transitions, the operator bases for Majorana and Dirac and the classification of lepton-number-violating/conserving interactions is applicable also for the study of $b\to s/d\nu\nu$ and $c\to u\nu\nu$ transitions relevant in current phenomenology., Comment: 23 pages, 6 figures

Visual Word Sense Disambiguation (VWSD) is a novel challenging task with the goal of retrieving an image among a set of candidates, which better represents the meaning of an ambiguous word within a given context. In this paper, we make a substantial step towards unveiling this interesting task by applying a varying set of approaches. Since VWSD is primarily a text-image retrieval task, we explore the latest transformer-based methods for multimodal retrieval. Additionally, we utilize Large Language Models (LLMs) as knowledge bases to enhance the given phrases and resolve ambiguity related to the target word. We also study VWSD as a unimodal problem by converting to text-to-text and image-to-image retrieval, as well as question-answering (QA), to fully explore the capabilities of relevant models. To tap into the implicit knowledge of LLMs, we experiment with Chain-of-Thought (CoT) prompting to guide explainable answer generation. On top of all, we train a learn to rank (LTR) model in order to combine our different modules, achieving competitive ranking results. Extensive experiments on VWSD demonstrate valuable insights to effectively drive future directions., Comment: Conference on Empirical Methods in Natural Language Processing (EMNLP) 2023

In this paper, we explore conjunctive query rewriting, focusing on queries containing universally quantified negation within the framework of disjunctive existential rules. We address the undecidability of the existence of a finite and complete UCQ-rewriting and the identification of finite unification sets (fus) of rules. We introduce new rule classes, connected linear rules and connected domain restricted rules, that exhibit the fus property for existential rules. Additionally, we propose disconnected disjunction for disjunctive existential rules to achieve the fus property when we extend the introduced rule fragments to disjunctive existential rules. We present ECOMPLETO, a system for efficient query rewriting with disjunctive existential rules, capable of handling UCQs with universally quantified negation. Our experiments demonstrate ECOMPLETO's consistent ability to produce finite UCQ-rewritings and describe the performance on different ontologies and queries., Comment: 23 pages, 4 figures

We investigate a mechanism of primordial black hole (PBH) formation that avoids any dependence on specific inflationary features or exotic physics. In this scenario, the required large curvature fluctuations leading to PBH formation are generated after inflation by the quantum fluctuations of a light stochastic spectator field during inflation, when this field transiently dominates the energy density. We calculate the dynamics of such a spectator field during and after inflation, the distribution of induced curvature perturbations and their non-Gaussian tails leading to the copious production of PBHs. For a plateau-like potential, this scenario produces an extended PBH mass distribution with a peak at the solar-mass scale when one takes into account the effects of the thermal history. What is remarkable in this scenario is the absence of parameter fine-tuning. Instead, it invokes an anthropic selection over all the realizations of PBH abundances predicted by the field stochasticity. This scenario offers a novel perspective for the formation of PBHs with minimal ingredients and without the need of fine-tuning. It is amenable to observational tests, notably with the gravitational-wave observations of black hole mergers and of a background at nanoHertz frequency, as recently observed by pulsar timing arrays., Comment: 11 pages, 7 figures

Visual Word Sense Disambiguation (VWSD) is a novel challenging task that lies between linguistic sense disambiguation and fine-grained multimodal retrieval. The recent advancements in the development of visiolinguistic (VL) transformers suggest some off-the-self implementations with encouraging results, which however we argue that can be further improved. To this end, we propose some knowledge-enhancement techniques towards improving the retrieval performance of VL transformers via the usage of Large Language Models (LLMs) as Knowledge Bases. More specifically, knowledge stored in LLMs is retrieved with the help of appropriate prompts in a zero-shot manner, achieving performance advancements. Moreover, we convert VWSD to a purely textual question-answering (QA) problem by considering generated image captions as multiple-choice candidate answers. Zero-shot and few-shot prompting strategies are leveraged to explore the potential of such a transformation, while Chain-of-Thought (CoT) prompting in the zero-shot setting is able to reveal the internal reasoning steps an LLM follows to select the appropriate candidate. In total, our presented approach is the first one to analyze the merits of exploiting knowledge stored in LLMs in different ways to solve WVSD.

Evidence for relativistic quantum criticality of antiferromagnetism and superconductivity in two-dimensional Dirac fermion systems has been found in large-scale quantum Monte Carlo simulations. However, the corresponding ($2+1$)-dimensional Gross--Neveu--Yukawa field theory with $N_f=2$ four-component Dirac fermions coupled to two triplets of order parameters does not exhibit a renormalization group fixed point at one-loop order. Instead, the theory only features a critical point for a large or very small fractional number of fermion flavors $N_f$, which disappears for a broad range of flavor numbers around the physical case, $N_f=2$, due to fixed-point annihilation. This raises the question on how to explain the observed scaling collapse in the quantum Monte Carlo data. Here, we extend previous renormalization-group analyses by studying a generalized model at two-loop order in $4-\epsilon$ spacetime dimensions. We determine the $\epsilon$ correction to the upper and lower critical flavor numbers for the fixed-point annihilation and find that they both go towards the physical case $N_f=2$. However, this only happens very slowly, such that an extrapolation to $\epsilon=1$ still suggests the absence of criticality in $2+1$ dimensions. Thereby, we consolidate the finding that the continuum field theory does not feature a stable renormalization-group fixed point and no true quantum criticality would be expected for the considered system. We briefly discuss a possible reconciliation in terms of a complex conformal field theory. Further, we also explore the fixed-point structure in an enlarged theory space and identify a candidate stable fixed-point solution., Comment: 13 pages, 6 figures, 2 tables

We investigate a family of four-dimensional quantum field theories with weakly interacting ultraviolet fixed points up to four loop order in perturbation theory. Key new ingredients are the three loop gauge contributions to quartic scalar beta functions, which we compute in the $\overline{\text{MS}}$ scheme for a template $SU(N_c)$ gauge theory coupled to $N_f$ fundamental fermions and elementary scalars. We then determine fixed point couplings, field and mass anomalous dimensions, and universal scaling exponents up to the first three non-trivial orders in a small Veneziano parameter. The phase diagram and UV-IR connecting trajectories are found and contrasted with asymptotic freedom. Further, the size of the conformal window, unitarity, and mechanisms leading to the loss of conformality are investigated. Our results provide blueprints for concrete 4d non-supersymmetric conformal field theories with standard model-like field content, and invite further model building., Comment: 17 pages, 2 external figures; v2: finite-N RGEs provided in appendix, minor adjustments, v3: new figure; match published version

Harnessing the power of Artificial Intelligence (AI) and m-health towards detecting new bio-markers indicative of the onset and progress of respiratory abnormalities/conditions has greatly attracted the scientific and research interest especially during COVID-19 pandemic. The smarty4covid dataset contains audio signals of cough (4,676), regular breathing (4,665), deep breathing (4,695) and voice (4,291) as recorded by means of mobile devices following a crowd-sourcing approach. Other self reported information is also included (e.g. COVID-19 virus tests), thus providing a comprehensive dataset for the development of COVID-19 risk detection models. The smarty4covid dataset is released in the form of a web-ontology language (OWL) knowledge base enabling data consolidation from other relevant datasets, complex queries and reasoning. It has been utilized towards the development of models able to: (i) extract clinically informative respiratory indicators from regular breathing records, and (ii) identify cough, breath and voice segments in crowd-sourced audio recordings. A new framework utilizing the smarty4covid OWL knowledge base towards generating counterfactual explanations in opaque AI-based COVID-19 risk detection models is proposed and validated., Comment: Submitted for publication in Nature Scientific Data

CP-odd Higgs couplings to bottom and charm quarks arise in many extensions of the standard model and are of potential interest for electroweak baryogenesis. These couplings induce a contribution to the electron EDM. The experimental limit on the latter then leads to a strong bound on the CP-odd Higgs couplings. We point out that this bound receives large QCD corrections, even though it arises from a leptonic observable. We calculate the contribution of CP-odd Higgs couplings to the bottom and charm quarks in renormalisation-group improved perturbation theory at next-to-leading order in the strong interaction, thereby reducing the uncertainty to a few percent., Comment: 21 pages, 4 figures, 1 table; updated numerics, extended discussion, added references. Version as published in JHEP

This paper describes the methodology followed and the lessons learned from employing crowdsourcing techniques as part of a homework assignment involving higher education students of computer science. Making use of a platform that supports crowdsourcing in the cultural heritage domain students were solicited to enrich the metadata associated with a selection of music tracks. The results of the campaign were further analyzed and exploited by students through the use of semantic web technologies. In total, 98 students participated in the campaign, contributing more than 6400 annotations concerning 854 tracks. The process also led to the creation of an openly available annotated dataset, which can be useful for machine learning models for music tagging. The campaign's results and the comments gathered through an online survey enable us to draw some useful insights about the benefits and challenges of integrating crowdsourcing into computer science curricula and how this can enhance students' engagement in the learning process., Comment: Presented in The 4th International Conference on Artificial Intelligence in Education Technology (AIET 2023), Berlin, Germany, 31 June-2 July 2023. For The GitHub code for the created music dataset, see https://github.com/vaslyb/MusicCrowd

In this paper we study the production of Primordial Black Holes (PBHs) from inflation in order to explain the Dark Mater (DM) in the Universe. The evaluation of the fractional PBHs abundance to DM is sensitive to the value of the threshold $\mathrm{\delta_c}$ and the exact value of $\mathrm{\delta_c}$ is sensitive to the specific shape of the cosmological fluctuations. Different mechanisms producing PBHs lead to different thresholds and hence to different fractional abundances of PBHs. In this study, we examine various classes of inflationary models proposed in the existing literature to elucidate the formation of PBHs and we evaluate numerically the associated threshold values. Having evaluated the thresholds we compute the abundances of PBHs to DM using the Press Schecter approach and the Peak Theory. Given the influence of different power spectra on the thresholds, we investigate whether these inflationary models can successfully account for a significant fraction of DM. Moreover, we provide suggested values for the critical threshold. By examining the interplay between inflationary models, threshold values, and PBH abundances, our study aims to shed light on the viability of PBHs as a candidate for DM and contributes to the ongoing discussion regarding the nature of DM in the Universe, Comment: 17 pages, 4 figures

Counterfactual explanations have been argued to be one of the most intuitive forms of explanation. They are typically defined as a minimal set of edits on a given data sample that, when applied, changes the output of a model on that sample. However, a minimal set of edits is not always clear and understandable to an end-user, as it could, for instance, constitute an adversarial example (which is indistinguishable from the original data sample to an end-user). Instead, there are recent ideas that the notion of minimality in the context of counterfactuals should refer to the semantics of the data sample, and not to the feature space. In this work, we build on these ideas, and propose a framework that provides counterfactual explanations in terms of knowledge graphs. We provide an algorithm for computing such explanations (given some assumptions about the underlying knowledge), and quantitatively evaluate the framework with a user study., Comment: To appear at IJCAI 2023

In the wake of responsible AI, interpretability methods, which attempt to provide an explanation for the predictions of neural models have seen rapid progress. In this work, we are concerned with explanations that are applicable to natural language processing (NLP) models and tasks, and we focus specifically on the analysis of counterfactual, contrastive explanations. We note that while there have been several explainers proposed to produce counterfactual explanations, their behaviour can vary significantly and the lack of a universal ground truth for the counterfactual edits imposes an insuperable barrier on their evaluation. We propose a new back translation-inspired evaluation methodology that utilises earlier outputs of the explainer as ground truth proxies to investigate the consistency of explainers. We show that by iteratively feeding the counterfactual to the explainer we can obtain valuable insights into the behaviour of both the predictor and the explainer models, and infer patterns that would be otherwise obscured. Using this methodology, we conduct a thorough analysis and propose a novel metric to evaluate the consistency of counterfactual generation approaches with different characteristics across available performance indicators., Comment: Accepted at Findings of ACL 2023