Search

Gravitational Wave (GW) detectors routinely encounter transient noise bursts, known as glitches, which are caused by either instrumental or environmental factors. Due to their high occurrence rate, glitches can overlap with GW signals, as in the notable case of GW170817, the first detection of a binary neutron star merger. Accurate reconstruction and subtraction of these glitches is a challenging problem that must be addressed to ensure that scientific conclusions drawn from the data are reliable. This problem will intensify with third-generation observatories like the Einstein Telescope (ET) due to their higher detection rates of GWs and the longer duration of signals within the sensitivity band of the detectors. Robust glitch mitigation algorithms are, therefore, crucial for maximizing the scientific output of next-generation GW observatories. For the first time, we demonstrate how the null stream inherent in ET's unique triangular configuration can be leveraged by state-of-the-art glitch characterization methodology to essentially undo the effect of glitches for the purpose of estimating the parameters of the source. The null stream based approach enables characterization and subtraction of glitches that occur arbitrarily close to the peak of the signal without any significant effect on the quality of parameter measurements, and achieves an order of magnitude computational speed-up compared to when the null stream is not available. By contrast, without the null stream, significant biases can occur in the glitch reconstruction, which deteriorate the quality of subsequent measurements of the source parameters. This demonstrates a clear edge which the null stream can offer for precision GW science in the ET era.

The concentration-virial mass (c-M) relation is a fundamental scaling relation within the standard cold dark matter ($\Lambda$CDM) framework well established in numerical simulations. However, observational constraints of this relation are hampered by the difficulty of characterising the properties of dark matter haloes. Recent comparisons between simulations and observations have suggested a systematic difference of the c-M relation, with higher concentrations in the latter. In this work, we undertake detailed comparisons between simulated galaxies and observations of a sample of strong-lensing galaxies. We explore several factors of the comparison with strong gravitational lensing constraints, including the choice of the generic dark matter density profile, the effect of radial resolution, the reconstruction limits of observed versus simulated mass profiles, and the role of the initial mass function in the derivation of the dark matter parameters. Furthermore, we show the dependence of the c-M relation on reconstruction and model errors through a detailed comparison of real and simulated gravitational lensing systems. An effective reconciliation of simulated and observed c-M relations can be achieved if one considers less strict assumptions on the dark matter profile, for example, by changing the slope of a generic NFW profile or focusing on rather extreme combinations of stellar-to-dark matter distributions. A minor effect is inherent to the applied method: fits to the NFW profile on a less well-constrained inner mass profile yield slightly higher concentrations and lower virial masses., Comment: 9 pages, 9 figures, 2 tables. Accepted for publication in A&A

Cosmological hydrodynamical simulations provide valuable insights on galaxy evolution when coupled with observational data. Comparisons with real galaxies are typically performed via scaling relations of model fitting. Here we follow an alternative approach based on the spectral variance in a model-independent way. We build upon the work presented in Sharbaf et al. (2023) that studied the covariance of high quality SDSS continuum-subtracted spectra in a relatively narrow range of velocity dispersion (100-150 km/s). Here the same analysis is applied to synthetic data from the EAGLE and Illustris TNG100 simulations, to assess the ability of these runs to mimic real galaxies. The real and simulated spectra are consistent regarding spectral variance, although with subtle differences that can inform the implementation of subgrid physics. Spectral fitting done a posteriori on stacks segregated with respect to latent space reveals that the first principal component (PC1) is predominantly influenced by the stellar age distribution, with an underlying age-metallicity degeneracy. Good agreement is found regarding star formation prescriptions but there is disagreement with AGN feedback, that also affects the subset of quiescent galaxies. We show a substantial difference in the implementation of the AGN subgrid, regarding central black hole seeding, that could lead to the mismatch. Differences are manifest between these two simulations in the star formation histories stacked with respect to latent space. We emphasise that this methodology only relies on the spectral variance to assess whether simulations provide a true representation of galaxy formation., Comment: Submitted to MNRAS. 27 pages including supplementary material. Comments welcome

This paper reports on the shared tasks organized by the 21st IWSLT Conference. The shared tasks address 7 scientific challenges in spoken language translation: simultaneous and offline translation, automatic subtitling and dubbing, speech-to-speech translation, dialect and low-resource speech translation, and Indic languages. The shared tasks attracted 18 teams whose submissions are documented in 26 system papers. The growing interest towards spoken language translation is also witnessed by the constantly increasing number of shared task organizers and contributors to the overview paper, almost evenly distributed across industry and academia., Comment: IWSLT 2024; 59 pages

Spurred by the demand for interpretable models, research on eXplainable AI for language technologies has experienced significant growth, with feature attribution methods emerging as a cornerstone of this progress. While prior work in NLP explored such methods for classification tasks and textual applications, explainability intersecting generation and speech is lagging, with existing techniques failing to account for the autoregressive nature of state-of-the-art models and to provide fine-grained, phonetically meaningful explanations. We address this gap by introducing Spectrogram Perturbation for Explainable Speech-to-text Generation (SPES), a feature attribution technique applicable to sequence generation tasks with autoregressive models. SPES provides explanations for each predicted token based on both the input spectrogram and the previously generated tokens. Extensive evaluation on speech recognition and translation demonstrates that SPES generates explanations that are faithful and plausible to humans.

A discussion on the readiness of Italian universities to address gender-related issues from a regional standpoint is proposed. A statistical analysis is conducted on data of all scholars enrolled in Italian universities from 2000 to 2023 to investigate why the glass ceiling of the full professor position remains so challenging to break in almost all scientific fields and across all regions of Italy.

Gender bias in machine translation (MT) is recognized as an issue that can harm people and society. And yet, advancements in the field rarely involve people, the final MT users, or inform how they might be impacted by biased technologies. Current evaluations are often restricted to automatic methods, which offer an opaque estimate of what the downstream impact of gender disparities might be. We conduct an extensive human-centered study to examine if and to what extent bias in MT brings harms with tangible costs, such as quality of service gaps across women and men. To this aim, we collect behavioral data from 90 participants, who post-edited MT outputs to ensure correct gender translation. Across multiple datasets, languages, and types of users, our study shows that feminine post-editing demands significantly more technical and temporal effort, also corresponding to higher financial costs. Existing bias measurements, however, fail to reflect the found disparities. Our findings advocate for human-centered approaches that can inform the societal impact of bias., Comment: Accepted ad EMNLP 2024

The rise of foundation models (FMs), coupled with regulatory efforts addressing their risks and impacts, has sparked significant interest in open-source models. However, existing speech FMs (SFMs) fall short of full compliance with the open-source principles, even if claimed otherwise, as no existing SFM has model weights, code, and training data publicly available under open-source terms. In this work, we take the first step toward filling this gap by focusing on the 24 official languages of the European Union (EU). We collect suitable training data by surveying automatic speech recognition datasets and unlabeled speech corpora under open-source compliant licenses, for a total of 950k hours. Additionally, we release automatic transcripts for 441k hours of unlabeled data under the permissive CC-BY license, thereby facilitating the creation of open-source SFMs for the EU languages., Comment: Accepted at EMNLP 2024 Main Conference

Transformers are one of the most successful architectures of modern neural networks. At their core there is the so-called attention mechanism, which recently interested the physics community as it can be written as the derivative of an energy function in certain cases: while it is possible to write the cross-attention layer as a modern Hopfield network, the same is not possible for the self-attention, which is used in the GPT architectures and other autoregressive models. In this work we show that it is possible to obtain the self-attention layer as the derivative of local energy terms, which resemble a pseudo-likelihood. We leverage the analogy with pseudo-likelihood to design a recurrent model that can be trained without backpropagation: the dynamics shows transient states that are strongly correlated with both train and test examples. Overall we present a novel framework to interpret self-attention as an attractor network, potentially paving the way for new theoretical approaches inspired from physics to understand transformers.

HERMES (High Energy Rapid Modular Ensemble of Satellites) Pathfinder is a space-borne mission based on a constellation of six nano-satellites flying in a low-Earth orbit (LEO). The 3U CubeSats, to be launched in early 2025, host miniaturized instruments with a hybrid Silicon Drift Detector/GAGG:Ce scintillator photodetector system, sensitive to X-rays and gamma-rays in a large energy band. HERMES will operate in conjunction with Australian Space Industry Responsive Intelligent Thermal (SpIRIT) 6U CubeSat, launched in December 2023. HERMES will probe the temporal emission of bright high-energy transients such as Gamma-Ray Bursts (GRBs), ensuring a fast transient localization in a field of view of several steradians exploiting the triangulation technique. HERMES intrinsically modular transient monitoring experiment represents a keystone capability to complement the next generation of gravitational wave experiments. In this paper we outline the scientific case, development and programmatic status of the mission, Comment: 13 pages, 16 figures. Proceedings of SPIE "Astronomical Telescopes and Instrumentation" 2024. arXiv admin note: text overlap with arXiv:2210.13866

We present Speech-MASSIVE, a multilingual Spoken Language Understanding (SLU) dataset comprising the speech counterpart for a portion of the MASSIVE textual corpus. Speech-MASSIVE covers 12 languages from different families and inherits from MASSIVE the annotations for the intent prediction and slot-filling tasks. Our extension is prompted by the scarcity of massively multilingual SLU datasets and the growing need for versatile speech datasets to assess foundation models (LLMs, speech encoders) across languages and tasks. We provide a multimodal, multitask, multilingual dataset and report SLU baselines using both cascaded and end-to-end architectures in various training scenarios (zero-shot, few-shot, and full fine-tune). Furthermore, we demonstrate the suitability of Speech-MASSIVE for benchmarking other tasks such as speech transcription, language identification, and speech translation. The dataset, models, and code are publicly available at: https://github.com/hlt-mt/Speech-MASSIVE, Comment: Accepted at INTERSPEECH 2024. This version includes the same content but with additional appendices

It has been recently shown that a learning transition happens when a Hopfield Network stores examples generated as superpositions of random features, where new attractors corresponding to such features appear in the model. In this work we reveal that the network also develops attractors corresponding to previously unseen examples generated with the same set of features. We explain this surprising behaviour in terms of spurious states of the learned features: we argue that, increasing the number of stored examples beyond the learning transition, the model also learns to mix the features to represent both stored and previously unseen examples. We support this claim with the computation of the phase diagram of the model.

The CUbesat Solar Polarimeter (CUSP) project is a future CubeSat mission orbiting the Earth aimed to measure the linear polarization of solar flares in the hard X-ray band, by means of a Compton scattering polarimeter. CUSP will allow us to study the magnetic reconnection and particle acceleration in the flaring magnetic structures of our star. The project is in the framework of the Italian Space Agency Alcor Program, which aims to develop new CubeSat missions. CUSP is approved for a Phase B study that will last for 12 months, starting in mid-2024. We report on the current status of the CUSP mission project as the outcome of the Phase A., Comment: Proceeding of SPIE Conference "Astronomical Telescopes+ Instrumentation", Yokohama (Japan), 16-21 June 2024. arXiv admin note: substantial text overlap with arXiv:2208.06211

The CUbesat Solar Polarimeter (CUSP) project aims to develop a constellation of two CubeSats orbiting the Earth to measure the linear polarization of solar flares in the hard X-ray band by means of a Compton scattering polarimeter on board of each satellite. CUSP will allow to study the magnetic reconnection and particle acceleration in the flaring magnetic structures. CUSP is a project approved for a Phase B study by the Italian Space Agency in the framework of the Alcor program aimed to develop CubeSat technologies and missions. In this paper we describe the a method for a multi-physical simulation analysis while analyzing some possible design optimization of the payload design solutions adopted. In particular, we report the mechanical design for each structural component, the results of static and dynamic finite element analysis, the preliminary thermo-mechanical analysis for two specific thermal cases (hot and cold orbit) and a topological optimization of the interface between the platform and the payload., Comment: Proceeding of SPIE Conference "Astronomical Telescopes+ Instrumentation", Yokohama (Japan), 16-21 June 2024

The CUbesat Solar Polarimeter (CUSP) project is a CubeSat mission orbiting the Earth aimed to measure the linear polarization of solar flares in the hard X-ray band by means of a Compton scattering polarimeter. CUSP will allow to study the magnetic reconnection and particle acceleration in the flaring magnetic structures of our star. CUSP is a project in the framework of the Alcor Program of the Italian Space Agency aimed to develop new CubeSat missions. It is approved for a Phase B study. In this work, we report on the accurate simulation of the detector's response to evaluate the scientific performance. A GEANT4 Monte Carlo simulation is used to assess the physical interactions of the source photons with the detector and the passive materials. Using this approach, we implemented a detailed CUSP Mass Model. In this work, we report on the evaluation of the detector's effective area as a function of the beam energy., Comment: Proceeding of SPIE Conference "Astronomical Telescopes+ Instrumentation", Yokohama (Japan), 16-21 June 2024

This paper describes the FBK's participation in the Simultaneous Translation Evaluation Campaign at IWSLT 2024. For this year's submission in the speech-to-text translation (ST) sub-track, we propose SimulSeamless, which is realized by combining AlignAtt and SeamlessM4T in its medium configuration. The SeamlessM4T model is used "off-the-shelf" and its simultaneous inference is enabled through the adoption of AlignAtt, a SimulST policy based on cross-attention that can be applied without any retraining or adaptation of the underlying model for the simultaneous task. We participated in all the Shared Task languages (English->{German, Japanese, Chinese}, and Czech->English), achieving acceptable or even better results compared to last year's submissions. SimulSeamless, covering more than 143 source languages and 200 target languages, is released at: https://github.com/hlt-mt/FBK-fairseq/.

Normalizing Flows (NFs) are powerful and efficient models for density estimation. When modeling densities on manifolds, NFs can be generalized to injective flows but the Jacobian determinant becomes computationally prohibitive. Current approaches either consider bounds on the log-likelihood or rely on some approximations of the Jacobian determinant. In contrast, we propose injective flows for star-like manifolds and show that for such manifolds we can compute the Jacobian determinant exactly and efficiently, with the same cost as NFs. This aspect is particularly relevant for variational inference settings, where no samples are available and only some unnormalized target is known. Among many, we showcase the relevance of modeling densities on star-like manifolds in two settings. Firstly, we introduce a novel Objective Bayesian approach for penalized likelihood models by interpreting level-sets of the penalty as star-like manifolds. Secondly, we consider probabilistic mixing models and introduce a general method for variational inference by defining the posterior of mixture weights on the probability simplex.

Streaming speech-to-text translation (StreamST) is the task of automatically translating speech while incrementally receiving an audio stream. Unlike simultaneous ST (SimulST), which deals with pre-segmented speech, StreamST faces the challenges of handling continuous and unbounded audio streams. This requires additional decisions about what to retain of the previous history, which is impractical to keep entirely due to latency and computational constraints. Despite the real-world demand for real-time ST, research on streaming translation remains limited, with existing works solely focusing on SimulST. To fill this gap, we introduce StreamAtt, the first StreamST policy, and propose StreamLAAL, the first StreamST latency metric designed to be comparable with existing metrics for SimulST. Extensive experiments across all 8 languages of MuST-C v1.0 show the effectiveness of StreamAtt compared to a naive streaming baseline and the related state-of-the-art SimulST policy, providing a first step in StreamST research., Comment: Accepted at ACL 2024 main conference

Human evaluation is a critical component in machine translation system development and has received much attention in text translation research. However, little prior work exists on the topic of human evaluation for speech translation, which adds additional challenges such as noisy data and segmentation mismatches. We take first steps to fill this gap by conducting a comprehensive human evaluation of the results of several shared tasks from the last International Workshop on Spoken Language Translation (IWSLT 2023). We propose an effective evaluation strategy based on automatic resegmentation and direct assessment with segment context. Our analysis revealed that: 1) the proposed evaluation strategy is robust and scores well-correlated with other types of human judgements; 2) automatic metrics are usually, but not always, well-correlated with direct assessment scores; and 3) COMET as a slightly stronger automatic metric than chrF, despite the segmentation noise introduced by the resegmentation step systems. We release the collected human-annotated data in order to encourage further investigation., Comment: LREC-COLING2024 publication (with corrections for Table 3)

Subtitling plays a crucial role in enhancing the accessibility of audiovisual content and encompasses three primary subtasks: translating spoken dialogue, segmenting translations into concise textual units, and estimating timestamps that govern their on-screen duration. Past attempts to automate this process rely, to varying degrees, on automatic transcripts, employed diversely for the three subtasks. In response to the acknowledged limitations associated with this reliance on transcripts, recent research has shifted towards transcription-free solutions for translation and segmentation, leaving the direct generation of timestamps as uncharted territory. To fill this gap, we introduce the first direct model capable of producing automatic subtitles, entirely eliminating any dependence on intermediate transcripts also for timestamp prediction. Experimental results, backed by manual evaluation, showcase our solution's new state-of-the-art performance across multiple language pairs and diverse conditions., Comment: Accepted to ACL 2024 main conference

To improve the storage capacity of the Hopfield model, we develop a version of the dreaming algorithm that perpetually reinforces the patterns to be stored (as in the Hebb rule), and erases the spurious memories (as in dreaming algorithms). For this reason, we called it Daydreaming. Daydreaming is not destructive and it converges asymptotically to stationary retrieval maps. When trained on random uncorrelated examples, the model shows optimal performance in terms of the size of the basins of attraction of stored examples and the quality of reconstruction. We also train the Daydreaming algorithm on correlated data obtained via the random-features model and argue that it spontaneously exploits the correlations thus increasing even further the storage capacity and the size of the basins of attraction. Moreover, the Daydreaming algorithm is also able to stabilize the features hidden in the data. Finally, we test Daydreaming on the MNIST dataset and show that it still works surprisingly well, producing attractors that are close to unseen examples and class prototypes.

Machine translation (MT) models are known to suffer from gender bias, especially when translating into languages with extensive gendered morphology. Accordingly, they still fall short in using gender-inclusive language, also representative of non-binary identities. In this paper, we look at gender-inclusive neomorphemes, neologistic elements that avoid binary gender markings as an approach towards fairer MT. In this direction, we explore prompting techniques with large language models (LLMs) to translate from English into Italian using neomorphemes. So far, this area has been under-explored due to its novelty and the lack of publicly available evaluation resources. We fill this gap by releasing Neo-GATE, a resource designed to evaluate gender-inclusive en-it translation with neomorphemes. With Neo-GATE, we assess four LLMs of different families and sizes and different prompt formats, identifying strengths and weaknesses of each on this novel task for MT., Comment: Accepted at EAMT 2024

The tremendous hype around autonomous driving is eagerly calling for emerging and novel technologies to support advanced mobility use cases. As car manufactures keep developing SAE level 3+ systems to improve the safety and comfort of passengers, traffic authorities need to establish new procedures to manage the transition from human-driven to fully-autonomous vehicles while providing a feedback-loop mechanism to fine-tune envisioned autonomous systems. Thus, a way to automatically profile autonomous vehicles and differentiate those from human-driven ones is a must. In this paper, we present a fully-fledged framework that monitors active vehicles using camera images and state information in order to determine whether vehicles are autonomous, without requiring any active notification from the vehicles themselves. Essentially, it builds on the cooperation among vehicles, which share their data acquired on the road feeding a machine learning model to identify autonomous cars. We extensively tested our solution and created the NexusStreet dataset, by means of the CARLA simulator, employing an autonomous driving control agent and a steering wheel maneuvered by licensed drivers. Experiments show it is possible to discriminate the two behaviors by analyzing video clips with an accuracy of 80%, which improves up to 93% when the target state information is available. Lastly, we deliberately degraded the state to observe how the framework performs under non-ideal data collection conditions.

The attention mechanism, a cornerstone of state-of-the-art neural models, faces computational hurdles in processing long sequences due to its quadratic complexity. Consequently, research efforts in the last few years focused on finding more efficient alternatives. Among them, Hyena (Poli et al., 2023) stands out for achieving competitive results in both language modeling and image classification, while offering sub-quadratic memory and computational complexity. Building on these promising results, we propose ConfHyena, a Conformer whose encoder self-attentions are replaced with an adaptation of Hyena for speech processing, where the long input sequences cause high computational costs. Through experiments in automatic speech recognition (for English) and translation (from English into 8 target languages), we show that our best ConfHyena model significantly reduces the training time by 27%, at the cost of minimal quality degradation (~1%), which, in most cases, is not statistically significant., Comment: Accepted at LREC-COLING 2024

The field of natural language processing (NLP) has recently witnessed a transformative shift with the emergence of foundation models, particularly Large Language Models (LLMs) that have revolutionized text-based NLP. This paradigm has extended to other modalities, including speech, where researchers are actively exploring the combination of Speech Foundation Models (SFMs) and LLMs into single, unified models capable of addressing multimodal tasks. Among such tasks, this paper focuses on speech-to-text translation (ST). By examining the published papers on the topic, we propose a unified view of the architectural solutions and training strategies presented so far, highlighting similarities and differences among them. Based on this examination, we not only organize the lessons learned but also show how diverse settings and evaluation approaches hinder the identification of the best-performing solution for each architectural building block and training choice. Lastly, we outline recommendations for future works on the topic aimed at better understanding the strengths and weaknesses of the SFM+LLM solutions for ST., Comment: Outstanding paper at the ACL 2024 main conference

Face recognition (FR) has reached a high technical maturity. However, its use needs to be carefully assessed from an ethical perspective, especially in sensitive scenarios. This is precisely the focus of this paper: the use of FR for the identification of specific subjects in moderately to densely crowded spaces (e.g. public spaces, sports stadiums, train stations) and law enforcement scenarios. In particular, there is a need to consider the trade-off between the need to protect privacy and fundamental rights of citizens as well as their safety. Recent Artificial Intelligence (AI) policies, notably the European AI Act, propose that such FR interventions should be proportionate and deployed only when strictly necessary. Nevertheless, concrete guidelines on how to address the concept of proportional FR intervention are lacking to date. This paper proposes a framework to contribute to assessing whether an FR intervention is proportionate or not for a given context of use in the above mentioned scenarios. It also identifies the main quantitative and qualitative variables relevant to the FR intervention decision (e.g. number of people in the scene, level of harm that the person(s) in search could perpetrate, consequences to individual rights and freedoms) and propose a 2D graphical model making it possible to balance these variables in terms of ethical cost vs security gain. Finally, different FR scenarios inspired by real-world deployments validate the proposed model. The framework is conceived as a simple support tool for decision makers when confronted with the deployment of an FR system., Comment: Published a shorter version in IEEE International Conference on Automatic Face and Gesture Recognition 2024

Gender-neutral translation (GNT) that avoids biased and undue binary assumptions is a pivotal challenge for the creation of more inclusive translation technologies. Advancements for this task in Machine Translation (MT), however, are hindered by the lack of dedicated parallel data, which are necessary to adapt MT systems to satisfy neutral constraints. For such a scenario, large language models offer hitherto unforeseen possibilities, as they come with the distinct advantage of being versatile in various (sub)tasks when provided with explicit instructions. In this paper, we explore this potential to automate GNT by comparing MT with the popular GPT-4 model. Through extensive manual analyses, our study empirically reveals the inherent limitations of current MT systems in generating GNTs and provides valuable insights into the potential and challenges associated with prompting for neutrality., Comment: Accepted at EACL 2024