Search

Videos are often used to learn or extract the necessary information to complete tasks in ways different than what text and static imagery alone can provide. However, many existing agent benchmarks neglect long-context video understanding, instead focusing on text or static image inputs. To bridge this gap, we introduce VideoWebArena (VideoWA), a benchmark for evaluating the capabilities of long-context multimodal agents for video understanding. VideoWA consists of 2,021 web agent tasks based on manually crafted video tutorials, which total almost four hours of content. For our benchmark, we define a taxonomy of long-context video-based agent tasks with two main areas of focus: skill retention and factual retention. While skill retention tasks evaluate whether an agent can use a given human demonstration to complete a task efficiently, the factual retention task evaluates whether an agent can retrieve instruction-relevant information from a video to complete a task. We find that the best model achieves 13.3% success on factual retention tasks and 45.8% on factual retention QA pairs, far below human performance at 73.9% and 79.3%, respectively. On skill retention tasks, long-context models perform worse with tutorials than without, exhibiting a 5% performance decrease in WebArena tasks and a 10.3% decrease in VisualWebArena tasks. Our work highlights the need to improve the agentic abilities of long-context multimodal models and provides a testbed for future development with long-context video agents.

In this work, we analyze the possibilities of certain gauge transformations regarding some specific spinorial dual structures. To this end, we define a general structure, which can be expressed in terms of discrete symmetry operators associated with parity, charge conjugation, and time reversal. To this end, we consider the demand of tensor covariance of the spinor bilinears under space-time transformations and also deep algebraic considerations such as the invertibility of the dual structure which guarantee that the physical information remains unchanged (respecting the Fierz-Pauli-Kofink identities), and so on. Later, we examine the global gauge freedom implicated by each choice of dual and point out the physical implications. The final goal is achieved by establishing a well-defined coupling between the Elko and standard model neutrinos in a prelude for a new formulation of the hypothetical so-called sterile neutrino., Comment: 7 pages, 2 Tables

Unlocking the potential of Industry 5.0 hinges on robust cybersecurity measures. This new Industrial Revolution prioritises human-centric values while addressing pressing societal issues such as resource conservation, climate change, and social stability. Recognising the heightened risk of cyberattacks due to the new enabling technologies in Industry 5.0, this paper analyses potential threats and corresponding countermeasures. Furthermore, it evaluates the existing industrial implementation frameworks, which reveals their inadequacy in ensuring a secure transition from Industry 4.0 to Industry 5.0. Consequently, the paper underscores the necessity of developing a new framework centred on cybersecurity to facilitate organisations' secure adoption of Industry 5.0 principles. The creation of such a framework is emphasised as a necessity for organisations., Comment: 7 pages, 2 figures, Accepted for PST2024 : 21st Annual International Conference on Privacy, Security, and Trust

Stage-IV galaxy surveys will measure correlations at small cosmological scales with high signal-to-noise ratio. One of the main challenges of extracting information from small scales is devising accurate models, as well as characterizing the theoretical uncertainties associated with any given model. In this work, we explore the mitigation of theoretical uncertainty due to nonlinear galaxy bias in the context of photometric 2$\times$2-pt analyses. We consider linear galaxy bias as the fiducial model and derive the contribution to the covariance matrix induced by neglected higher-order bias. We construct a covariance matrix for the theoretical error in galaxy clustering and galaxy-galaxy lensing using simulation-based relations that connect higher-order parameters to linear bias. We test the modified likelihood in 2$\times$2-pt analyses based on two sets of mock data vectors: (1) simulated data vectors, constructed from those same relations between bias parameters, and (2) data vectors based on the AbacusSummit simulation suite. We then compare the performance of the theoretical-error approach to the commonly employed scale cuts. We find most theoretical-error configurations yield results equivalent to the scale cuts in terms of precision and accuracy, in some cases producing significantly stronger bounds on cosmological parameters. These results are independent of the maximum scale $k_\mathrm{max}$ in the analysis with theoretical error. The scenarios where linear bias supplemented by theoretical error is unable to recover unbiased cosmology are connected to inadequate modeling of the $gg$-$g\kappa$ covariance of theoretical error. In view of its removing the ambiguity in the choice of $k_\mathrm{max}$, as well as the possibility of attaining higher precision than the usual scale cuts, we consider this method to be promising for analyses of LSS in upcoming photometric galaxy surveys., Comment: 30 pages, 9 figures; references added; submitted to JCAP

We use James Webb Space Telescope (JWST) Mid-Infrared Instrument (MIRI) medium-resolution spectrometer (MRS) observations of UGC 8782, CGCG 012-070 and NGC 3884 to investigate the origin of the H$_2$ emission. These three nearby AGN hosts are known to present H$_2$ emission excess relative to star-forming galaxies, as traced by the H$_2$ S(3)/PAH$_{\rm 11.3\mu m}$ line ratio. For every spaxel, we also measure the velocity width ($W_{\rm 80}$) of H$_2$ S(3). We find that the distribution of H$_2$ in the space of these parameters - H$_2$/PAH and its velocity width - is bimodal: one cluster of points shows approximately constant values of $W_{\rm 80}$, while the other cluster exhibits a strong correlation between $W_{\rm 80}$ and H$_2$/PAH. We estimated the temperature of the H$_2$ gas assuming a power-law distribution and find flatter distributions in regions where a correlation between $W_{\rm 80}$ and H$_2$ S(3)/PAH$_{\rm 11.3\mu m}$ is observed. Additionally, we observed a correlation between the [Fe II]$_{\rm 5.34 \mu m}$ (a known shock tracer) and the H$_2$ emission. This indicates that the excess H$_2$ emission excess is associated to shock heating of the gas, generated by outflows or by the interaction of the radio jet with the ambient gas., Comment: 12 pages, 5 figures, submitted to ApJ. Comments are welcome

The influence of Active Galactic Nuclei (AGN) on star formation within their host galaxies remains a topic of intense debate. One of the primary challenges in quantifying the star formation rate (SFR) within AGN hosts arises from the prevalent assumption in most methodologies, which attribute gas excitation to young stars alone. However, this assumption does not consider the contribution of the AGN to the ionization of the gas in their environment. To address this issue, we evaluate the use of strong optical emission lines to obtain the SFR surface density ($\Sigma{{\rm SFR_{AGN}}}$) in regions predominantly ionized by an AGN, using a sample of 293 AGN hosts from the MaNGA survey, with SFR measurements available through stellar population fitting. We propose calibrations involving the H$\alpha$ and [O\,{\sc iii}]$\lambda$5007 emission lines, which can be used to determine $\Sigma{{\rm SFR_{AGN}}}$, resulting in values consistent with those estimated through stellar population fitting., Comment: Accepted for publication in MNRAS

In this work, we propose a novel method (GLOV) enabling Large Language Models (LLMs) to act as implicit Optimizers for Vision-Langugage Models (VLMs) to enhance downstream vision tasks. Our GLOV meta-prompts an LLM with the downstream task description, querying it for suitable VLM prompts (e.g., for zero-shot classification with CLIP). These prompts are ranked according to a purity measure obtained through a fitness function. In each respective optimization step, the ranked prompts are fed as in-context examples (with their accuracies) to equip the LLM with the knowledge of the type of text prompts preferred by the downstream VLM. Furthermore, we also explicitly steer the LLM generation process in each optimization step by specifically adding an offset difference vector of the embeddings from the positive and negative solutions found by the LLM, in previous optimization steps, to the intermediate layer of the network for the next generation step. This offset vector steers the LLM generation toward the type of language preferred by the downstream VLM, resulting in enhanced performance on the downstream vision tasks. We comprehensively evaluate our GLOV on 16 diverse datasets using two families of VLMs, i.e., dual-encoder (e.g., CLIP) and encoder-decoder (e.g., LLaVa) models -- showing that the discovered solutions can enhance the recognition performance by up to 15.0% and 57.5% (3.8% and 21.6% on average) for these models., Comment: Code: https://github.com/jmiemirza/GLOV

The near-axis description of optimised stellarators, at second order in the expansion, provides important information about the field, both of physical and practical importance for stellarator optimisation. It however remains relatively underdeveloped for an important class of such stellarators, called quasi-isodynamic (QI). In this paper we develop the theoretical and numerical framework for the construction of such solutions. We find that the case of QI stellarators calls for the careful treatment of continuity, smoothness and periodicity of the various functions involved, especially for so-called half-helicity fields, which feature prominently in existing QI designs. The numerical implementation of necessary elements is described, and several examples are constructed and quantitatively verified in detail. This work establishes a basis for further systematic exploration of the space of QI stellarators, and the development of both theoretical and practical tools to facilitate effective optimisation of QI stellarators.

We explore the relations between the gas-phase metallicity radial profiles (few hundred inner parsec) and multiple galaxy properties for 15 Seyfert galaxies from the AGNIFS (Active Galactic Nuclei Integral Field Spectroscopy) sample using optical Integral Field Unit (IFU) observations from Gemini Multi-Object Spectrographs (GMOS) and Multi Unit Spectroscopic Explorer (MUSE) processed archival data. The data were selected at $z \lesssim 0.013$ within black hole mass range $\left[6<\log \left(M_{\rm BH}/{\rm M_\odot} \right)<9\right]$ with moderate 14--150\,keV X-ray luminosities $\left[42\,\lesssim\,\log L_X (\rm erg\,s^{-1})\,\lesssim\,44\right]$. We estimated the gas-phase metallicity using the strong-line methods and found mean values for the oxygen dependent ($Z \sim 0.75Z_\odot$) and nitrogen dependent ($Z \sim 1.14Z_\odot$) calibrations. These estimates show excellent agreement with $\Delta Z \approx 0.19$ dex and $\Delta Z \approx 0.18$ dex between the mean values from the two strong-line calibrations for GMOS and MUSE respectively, consistent with the order of metallicity uncertainty via the strong-line methods. We contend that our findings align with a scenario wherein local Seyferts have undergone seamless gas accretion histories, resulting in positive metallicity profile over an extended period of time, thereby providing insights into galaxy evolution and the chemical enrichment or depletion of the universe. Additionally, we argue that metal-poor gas inflow from the local interstellar medium (ISM) and accreted through the circumgalactic medium (CGM) onto the galaxy systems regulates the star formation processes by diluting their central metallicity and inverting their metallicity gradients, producing a more prominent anti-correlation between gas-phase metallicity and Eddington ratio., Comment: Accepted for publication in MNRAS; doi: 10.1093/mnras/stae2263

The advent of large integral field spectroscopic surveys has found that elliptical galaxies (EGs) can be classified into two classes: the fast rotators (whose kinematics are dominated by rotation) and the slow rotators (which exhibit slow or no rotation pattern). It is often suggested that while the slow rotators typically have boxy isophotal shapes, have a high $\alpha$-to-iron abundance ratio, and are quite massive, the fast rotators often exhibit the opposite properties (that is, having disky isophotes, lower $\alpha$-to-iron ratio, and of typical masses). Whether the EGs consist of two distinct populations (i.e., a dichotomy exists), remains an unsolved issue. To examine the existence of the dichotomy, we used a sample of 1,895 EGs from the SDSS-IV MaNGA survey, and measured robustly the stellar kinematics, isophotal shapes, and [Mg/Fe] ratio. We confirmed the previous finding that the bulk of the EGs are disky (65%) and fast rotators (67%), but found no evidence supporting a dichotomy, based on a principal component analysis. The different classes (boxy/disky and slow/fast rotators) of EGs occupy slightly different loci in the principal component space. This may explain the observed trends that led to the premature support of a dichotomy based on small samples of galaxies., Comment: 27 pages, 19 figures, 5 tables. Submitted to ApJ. Comments are welcome!

This work is dedicated to debias the Near-Earth Objects (NEO) population based on observations from the Asteroid Terrestrial-impact Last Alert System (ATLAS) telescopes. We have applied similar methods used to develop the recently released NEO model generator (NEOMOD), once debiasing the NEO population using data from Catalina Sky Survey (CSS) G96 telescope. ATLAS is composed of four different telescopes. We first analyzed observational data from each of all four telescopes separately and later combined them. Our results highlight main differences between CSS and ATLAS, e.g., sky coverage and survey power at debiasing the NEO population. ATLAS has a much larger sky coverage than CSS, allowing it to find bright NEOs that would be constantly "hiding" from CSS. Consequently, ATLAS is more powerful than CSS at debiasing the NEO population for H $\lesssim$ 19. With its intrinsically greater sensitivity and emphasis on observing near opposition, CSS excels in the debiasing of smaller objects. ATLAS, as an all sky survey designed to find imminent hazardous objects, necessarily spends a significant fraction of time looking at places on the sky where objects do not appear, reducing its power for debiasing the population of small objects. We estimate a NEO population completeness of $\approx$ 88%$^{+3\%}_{-2\%}$ for H $<$ 17.75 and $\approx$ 36%$^{+1\%}_{-1\%}$ for H $<$ 22.25. Those numbers are similar to previous estimates (within error bars for H $<$ 17.75) from CSS, yet, around 3% and 8% smaller at their face values, respectively. We also confirm previous finding that the $\nu_6$ secular resonance is the main source of small and faint NEOs at H = 28, whereas the 3:1 mean motion resonance with Jupiter dominates for larger and brighter NEOs at H = 15., Comment: Accepted for publication in Icarus, 26 pages, 13 figures, 2 tables

Large language models (LLMs) show remarkable potential to act as computer agents, enhancing human productivity and software accessibility in multi-modal tasks that require planning and reasoning. However, measuring agent performance in realistic environments remains a challenge since: (i) most benchmarks are limited to specific modalities or domains (e.g. text-only, web navigation, Q&A, coding) and (ii) full benchmark evaluations are slow (on order of magnitude of days) given the multi-step sequential nature of tasks. To address these challenges, we introduce the Windows Agent Arena: a reproducible, general environment focusing exclusively on the Windows operating system (OS) where agents can operate freely within a real Windows OS and use the same wide range of applications, tools, and web browsers available to human users when solving tasks. We adapt the OSWorld framework (Xie et al., 2024) to create 150+ diverse Windows tasks across representative domains that require agent abilities in planning, screen understanding, and tool usage. Our benchmark is scalable and can be seamlessly parallelized in Azure for a full benchmark evaluation in as little as 20 minutes. To demonstrate Windows Agent Arena's capabilities, we also introduce a new multi-modal agent, Navi. Our agent achieves a success rate of 19.5% in the Windows domain, compared to 74.5% performance of an unassisted human. Navi also demonstrates strong performance on another popular web-based benchmark, Mind2Web. We offer extensive quantitative and qualitative analysis of Navi's performance, and provide insights into the opportunities for future research in agent development and data generation using Windows Agent Arena. Webpage: https://microsoft.github.io/WindowsAgentArena Code: https://github.com/microsoft/WindowsAgentArena

We studied the ionized gas in the inner region ($\sim680\times470$ pc$^2$) of the galaxy NGC 6868 using Gemini/GMOS integral field unit observations. Channel maps reveal complex kinematics and morphology, indicating multiple processes at work in NGC 6868. Through emission-line fitting, we identified two ubiquitous components in our data: a narrow ($\sigma\sim110$ km s$^{-1}$) tracing an ionized gas disc and a broad component ($\sigma\sim300$ km s$^{-1}$) mainly associated with inflowing/outflowing gas. The derived V-band reddening shows a spatial distribution consistent with that obtained from stellar population synthesis, although with generally higher values. For the first time, we measured the electron temperature in NGC 6868, finding values ranging from $\sim 14000$ K in the central region to $\gtrsim20000$ K with an outward increasing temperature gradient. The electron density map exhibits an inverse relationship, with central values reaching $N_e\sim4000$ cm$^{-3}$ for the broad component decreasing to $N_e\sim100$ cm$^{-3}$ towards the edges of the field of view. Using BPT diagrams, we found that all spaxels are consistent with both AGN and shock ionization. However, when this information is combined with our kinematic and temperature findings, and further supported by the WHAN diagram, we argue that an AGN is the dominant ionisation mechanism in the central region of NGC 6868, while the extended outer component is ionized by a combination of hot low-mass evolved stars and shocks. According to our findings, shocks play a significant role in the ionization balance of this galaxy., Comment: 16 pages, 16 figures, accepted for publication in MNRAS

In this paper we consider block languages, namely sets of words having the same length, and study the deterministic and nondeterministic state complexity of several operations on these languages. Being a subclass of finite languages, the upper bounds of operational state complexity known for finite languages apply for block languages as well. However, in several cases, smaller values were found. Block languages can be represented as bitmaps, which are a good tool to study their minimal finite automata and their operations, as we illustrate here., Comment: In Proceedings NCMA 2024, arXiv:2409.06120

This is an index of the contributions by the Giant Radio Array for Neutrino Detection (GRAND) Collaboration to the 10th International Workshop on Acoustic and Radio EeV Neutrino Detection Activities (ARENA 2024, University of Chicago, June 11-14, 2024). The contributions include an overview of GRAND in its present and future incarnations, methods of radio-detection that are being developed for them, and ongoing joint work between the GRAND and BEACON experiments., Comment: Note: To access the list of contributions, please follow the "HTML" link that can be found on the arXiv page

In this work, we shall analyze the necessity of a proper definition of the dual structure for Elko spinors, and singular spinors in general. We examine in detail why the Dirac dual structure fails and it is not functional for these cases, highlighting all the physical consequences of this misdirection. The approach considered here is different from the one usually taken in current literature. We pinpoint the shortcomings that this dual structure brings to particle interpretation, and propagator structure, as well as the implications in non-locality, and the existence of negative energy levels. This investigation furnishes a background for the successful achievements in the field associated with such proper dual formulation., Comment: 10 pages

Baryonic Acoustic Oscillation (BAO) data from the Dark Energy Spectroscopic Instrument (DESI), in combination with Cosmic Microwave Background (CMB) data and Type Ia Supernovae (SN) luminosity distances, suggests a dynamical evolution of the dark energy equation of state with a phantom phase ($w < -1$) in the past when the so-called $w_0w_a$ parametrization $w(a) = w_0 + w_a(1-a)$ is assumed. In this work, we investigate more general dark energy models that also allow a phantom equation of state. We consider three cases: an equation of state with a transition feature, a model-agnostic equation of state with constant values in chosen redshift bins, and a k-essence model. Since the dark energy equation of state is correlated with neutrino masses, we reassess constraints on the neutrino mass sum focusing on the model-agnostic equation of state. We find that the combination of DESI BAO with Planck 2018 CMB data and SN data from Pantheon, Pantheon+, or Union3 is consistent with an oscillatory dark energy equation of state, while a monotonic behavior is preferred by the DESY5 SN data. Performing model comparison techniques, we find that the $w_0w_a$ parametrization remains the simplest dark energy model that can provide a better fit to DESI BAO, CMB, and all SN datasets than $\Lambda$CDM. Constraints on the neutrino mass sum assuming dynamical dark energy are relaxed compared to $\Lambda$CDM and we show that these constraints are tighter in the model-agnostic case relative to $w_0w_a$ model by $70\%-90\%$., Comment: 36 pages, 12 figures, 4 tables

The operation of upcoming ultra-high-energy cosmic-ray, gamma-ray, and neutrino radio-detection experiments, like the Giant Radio Array for Neutrino Detection (GRAND), poses significant computational challenges involving the production of numerous simulations of particle showers and their detection, and a high data throughput. GRANDlib is an open-source software tool designed to meet these challenges. Its primary goal is to perform end-to-end simulations of the detector operation, from the interaction of ultra-high-energy particles, through -- by interfacing with external air-shower simulations -- the ensuing particle shower development and its radio emission, to its detection by antenna arrays and its processing by data-acquisition systems. Additionally, GRANDlib manages the visualization, storage, and retrieval of experimental and simulated data. We present an overview of GRANDlib to serve as the basis of future GRAND analyses., Comment: 11 pages, 9 figures, plus appendices

We use the James Webb Space Telescope (JWST) Mid-Infrared Instrument (MIRI) medium-resolution spectrometer (MRS) observations of the radio loud AGN host UGC 8782 to map the warm molecular and ionized gas kinematics. The data reveal spatially resolved outflows in the inner 2 kpc, seen in low ionization (traced by the [Ar ii] 6.99um emission) and in warm molecular gas (traced by the H2 rotational transitions). We find a maximum mass-outflow rate of 4.90 (2.04) M/yr at ~900 pc from the nucleus for the warm outflow (198 K< T< 1000 K) and estimate and outflow rate of up to 1.22 (0.51) M/yr for the hotter gas phase (T > 1000 K). These outflows can clear the entire nuclear reservoir of warm molecular gas in about 1 Myr. The derived kinetic power of the molecular outflows lead to coupling efficiencies of 2-5 percent of the AGN luminosity, way above the minimum expected to the AGN feedback be effective quenching the star formation., Comment: Accepted for publication at APJ

The paradigm of Big Data has been established as a solid field of studies in many areas such as healthcare, science, transport, education, government services, among others. Despite widely discussed, there is no agreed definition about the paradigm although there are many concepts proposed by the academy and industry. This work aims to provide an analytical view of the studies conducted and published regarding the Big Data paradigm. The approach used is the systematic map of the literature, combining bibliometric analysis and content analysis to depict the panorama of research works, identifying patterns, trends, and gaps. The results indicate that there is still a long way to go, both in research and in concepts, such as building and defining adequate infrastructures and standards, to meet future challenges and for the paradigm to become effective and bring the expected benefits., Comment: 8 pages, 1 figure, 5 tables

In this paper we study germs of holomorphic foliations, at the origin of the complex plane, tangent to Pfaffian hypersurfaces - integral hypersurfaces of real analytic 1-forms - satisfying the Rolle-Khovanskii condition. This hypothesis leads us to conclude that such a foliation is defined by a closed meromorphic 1-form, also allowing the classification of the simple models in its reduction of singularities., Comment: 23 pages

The quasiparticle density observed in low-temperature superconducting circuits is several orders of magnitude higher than the value expected at thermal equilibrium. The tunneling of this excess of quasiparticles across Josephson junctions is recognized as one of the main loss and decoherence mechanisms in superconducting qubits. Here we propose an additional loss mechanism arising from nonequilibrium quasiparticle densities: Ohmic loss due to quasiparticles residing in superconducting wires away from the junctions. Our theory leverages the recent experimental demonstration that the excess quasiparticles are in quasiequilibrium [T. Connolly et al., $\href{https://doi.org/10.1103/PhysRevLett.132.217001}{Phys. Rev. Lett. {\bf 132}, 217001 (2024)}$] and uses a generalized fluctuation-dissipation theorem to predict the amount of charge and flux noise generated by them. We show that the resulting charge noise can be larger than dielectric loss due to amorphous two-level systems at frequencies in the MHz range, and find a logarithmic-in-frequency ``nearly white" contribution to flux noise that is comparable to the flux noise observed in experiments. This shows that wire-resident quasiparticles are a universal source of loss and decoherence even when the quasiparticles are far away from Josephson junctions., Comment: 7 pages, 3 figures

This paper introduces long-context Granite code models that support effective context windows of up to 128K tokens. Our solution for scaling context length of Granite 3B/8B code models from 2K/4K to 128K consists of a light-weight continual pretraining by gradually increasing its RoPE base frequency with repository-level file packing and length-upsampled long-context data. Additionally, we also release instruction-tuned models with long-context support which are derived by further finetuning the long context base models on a mix of permissively licensed short and long-context instruction-response pairs. While comparing to the original short-context Granite code models, our long-context models achieve significant improvements on long-context tasks without any noticeable performance degradation on regular code completion benchmarks (e.g., HumanEval). We release all our long-context Granite code models under an Apache 2.0 license for both research and commercial use.

State-space models (SSMs) have emerged as an alternative to Transformers for audio modeling due to their high computational efficiency with long inputs. While recent efforts on Audio SSMs have reported encouraging results, two main limitations remain: First, in 10-second short audio tagging tasks, Audio SSMs still underperform compared to Transformer-based models such as Audio Spectrogram Transformer (AST). Second, although Audio SSMs theoretically support long audio inputs, their actual performance with long audio has not been thoroughly evaluated. To address these limitations, in this paper, 1) We applied knowledge distillation in audio space model training, resulting in a model called Knowledge Distilled Audio SSM (DASS). To the best of our knowledge, it is the first SSM that outperforms the Transformers on AudioSet and achieves an mAP of 47.6; and 2) We designed a new test called Audio Needle In A Haystack (Audio NIAH). We find that DASS, trained with only 10-second audio clips, can retrieve sound events in audio recordings up to 2.5 hours long, while the AST model fails when the input is just 50 seconds, demonstrating SSMs are indeed more duration scalable.

We discuss the motion of electrically and magnetically charged particles in the electromagnetic swirling universe. We show that the equations of motion can be decoupled in the Hamilton-Jacobi formalism, revealing the existence of a fourth constant of motion. The equations of motion can be analytically integrated. The solutions are presented in terms of elementary and elliptic functions. In addition, we discuss the possible orbits for both uncharged particles (in which case the motion is geodesic) and charged particles, respectively. A typical orbit is bounded in the radial direction and escapes to infinity in the $z-$ direction. However, the presence of the electromagnetic fields also leads to the existence of planar orbits., Comment: 14 pages, 3 figures

Using Large Language Models (LLMs) to generate synthetic data for model training has become increasingly popular in recent years. While LLMs are capable of producing realistic training data, the effectiveness of data generation is influenced by various factors, including the choice of prompt, task complexity, and the quality, quantity, and diversity of the generated data. In this work, we focus exclusively on using synthetic data for text classification tasks. Specifically, we use natural language understanding (NLU) models trained on synthetic data to assess the quality of synthetic data from different generation approaches. This work provides an empirical analysis of the impact of these factors and offers recommendations for better data generation practices., Comment: Accepted by DMLR @ ICML 2024

In this paper, we provide a set of Hermitian interactions for quantum fields based on Elko, considering the recent achievements concerning the most general form of singular spinors and Wigner degeneracy. We consider Hermiticity and renormalizability a criterion to define the derivative Elko-Higgs interaction as the suitable candidate for a dark coupling. Then, the free parameters of the model are fixed by cosmological constraints on the dark matter abundance, bounds on the Higgs invisible decays, and limits on the electron-dark matter scattering mediated by the Higgs., Comment: 17 pages, 2 figures

Recently, Large Language Models (LLMs) attained impressive performance in math and reasoning benchmarks. However, they still often struggle with logic problems and puzzles that are relatively easy for humans. To further investigate this, we introduce a new benchmark, SearchBench, containing 11 unique search problem types, each equipped with automated pipelines to generate an arbitrary number of instances and analyze the feasibility, correctness, and optimality of LLM-generated solutions. We show that even the most advanced LLMs fail to solve these problems end-to-end in text, e.g. GPT4 solves only 1.4%. SearchBench problems require considering multiple pathways to the solution as well as backtracking, posing a significant challenge to auto-regressive models. Instructing LLMs to generate code that solves the problem helps, but only slightly, e.g., GPT4's performance rises to 11.7%. In this work, we show that in-context learning with A* algorithm implementations enhances performance. The full potential of this promoting approach emerges when combined with our proposed Multi-Stage-Multi-Try method, which breaks down the algorithm implementation into two stages and verifies the first stage against unit tests, raising GPT-4's performance above 57%.

We present Self-MoE, an approach that transforms a monolithic LLM into a compositional, modular system of self-specialized experts, named MiXSE (MiXture of Self-specialized Experts). Our approach leverages self-specialization, which constructs expert modules using self-generated synthetic data, each equipping a shared base LLM with distinct domain-specific capabilities, activated via self-optimized routing. This allows for dynamic and capability-specific handling of various target tasks, enhancing overall capabilities, without extensive human-labeled data and added parameters. Our empirical results reveal that specializing LLMs may exhibit potential trade-offs in performances on non-specialized tasks. On the other hand, our Self-MoE demonstrates substantial improvements (6.5%p on average) over the base LLM across diverse benchmarks such as knowledge, reasoning, math, and coding. It also consistently outperforms other methods, including instance merging and weight merging, while offering better flexibility and interpretability by design with semantic experts and routing. Our findings highlight the critical role of modularity, the applicability of Self-MoE to multiple base LLMs, and the potential of self-improvement in achieving efficient, scalable, and adaptable systems.

Audio-Visual Speech Recognition (AVSR) uses lip-based video to improve performance in noise. Since videos are harder to obtain than audio, the video training data of AVSR models is usually limited to a few thousand hours. In contrast, speech models such as Whisper are trained with hundreds of thousands of hours of data, and thus learn a better speech-to-text decoder. The huge training data difference motivates us to adapt Whisper to handle video inputs. Inspired by Flamingo which injects visual features into language models, we propose Whisper-Flamingo which integrates visual features into the Whisper speech recognition and translation model with gated cross attention. Our audio-visual Whisper-Flamingo outperforms audio-only Whisper on English speech recognition and En-X translation for 6 languages in noisy conditions. Moreover, Whisper-Flamingo is a versatile model and conducts all of these tasks using one set of parameters, while prior methods are trained separately on each language., Comment: Interspeech 2024. Code https://github.com/roudimit/whisper-flamingo

Comparing two images in terms of Commonalities and Differences (CaD) is a fundamental human capability that forms the basis of advanced visual reasoning and interpretation. It is essential for the generation of detailed and contextually relevant descriptions, performing comparative analysis, novelty detection, and making informed decisions based on visual data. However, surprisingly, little attention has been given to these fundamental concepts in the best current mimic of human visual intelligence - Large Multimodal Models (LMMs). We develop and contribute a new two-phase approach CaD-VI for collecting synthetic visual instructions, together with an instruction-following dataset CaD-Inst containing 349K image pairs with CaD instructions collected using CaD-VI. Our approach significantly improves the CaD spotting capabilities in LMMs, advancing the SOTA on a diverse set of related tasks by up to 17.5%. It is also complementary to existing difference-only instruction datasets, allowing automatic targeted refinement of those resources increasing their effectiveness for CaD tuning by up to 10%. Additionally, we propose an evaluation benchmark with 7.5K open-ended QAs to assess the CaD understanding abilities of LMMs., Comment: Project page: https://wlin-at.github.io/cad_vi ; Huggingface dataset repo: https://huggingface.co/datasets/wlin21at/CaD-Inst

Compositional Reasoning (CR) entails grasping the significance of attributes, relations, and word order. Recent Vision-Language Models (VLMs), comprising a visual encoder and a Large Language Model (LLM) decoder, have demonstrated remarkable proficiency in such reasoning tasks. This prompts a crucial question: have VLMs effectively tackled the CR challenge? We conjecture that existing CR benchmarks may not adequately push the boundaries of modern VLMs due to the reliance on an LLM-only negative text generation pipeline. Consequently, the negatives produced either appear as outliers from the natural language distribution learned by VLMs' LLM decoders or as improbable within the corresponding image context. To address these limitations, we introduce ConMe -- a compositional reasoning benchmark and a novel data generation pipeline leveraging VLMs to produce `hard CR Q&A'. Through a new concept of VLMs conversing with each other to collaboratively expose their weaknesses, our pipeline autonomously generates, evaluates, and selects challenging compositional reasoning questions, establishing a robust CR benchmark, also subsequently validated manually. Our benchmark provokes a noteworthy, up to 33%, decrease in CR performance compared to preceding benchmarks, reinstating the CR challenge even for state-of-the-art VLMs., Comment: The first three authors contributed equally

We study 51 jellyfish galaxy candidates in the Fornax, Antlia, and Hydra clusters. These candidates are identified using the JClass scheme based on the visual classification of wide-field, twelve-band optical images obtained from the Southern Photometric Local Universe Survey. A comprehensive astrophysical analysis of the jellyfish (JClass > 0), non-jellyfish (JClass = 0), and independently organized control samples is undertaken. We develop a semi-automated pipeline using self-supervised learning and similarity search to detect jellyfish galaxies. The proposed framework is designed to assist visual classifiers by providing more reliable JClasses for galaxies. We find that jellyfish candidates exhibit a lower Gini coefficient, higher entropy, and a lower 2D S\'ersic index as the jellyfish features in these galaxies become more pronounced. Jellyfish candidates show elevated star formation rates (including contributions from the main body and tails) by $\sim$1.75 dex, suggesting a significant increase in the SFR caused by the ram-pressure stripping phenomenon. Galaxies in the Antlia and Fornax clusters preferentially fall towards the cluster's centre, whereas only a mild preference is observed for Hydra galaxies. Our self-supervised pipeline, applied in visually challenging cases, offers two main advantages: it reduces human visual biases and scales effectively for large datasets. This versatile framework promises substantial enhancements in morphology studies for future galaxy image surveys., Comment: 25 pages, 22 figures, accepted for publication in MNRAS

Abell 1758 (z~0.278) is a galaxy cluster composed of two structures: A1758N and A1758S, separated by ~2.2 Mpc. The northern cluster is itself a dissociative merging cluster that has already been modelled by dedicated simulations. Recent radio observations revealed the existence of a previously undetected bridge connecting A1758N and A1758S. New simulations are now needed to take into account the presence of A1758S. We wish to evaluate which orbital configuration would be compatible with a bridge between the clusters. Using N-body hydrodynamical simulations that build upon the previous model, we explore different scenarios that could have led to the current observed configuration. Five types of orbital approaches were tested: radial, tangential, vertical, post-apocentric, and outgoing. We found that the incoming simulated scenarios are generally consistent with mild enhancements of gas density between the approaching clusters. The mock X-ray images exhibit a detectable bridge in all cases. Compared to measurements of Chandra data, the amplitude of the X-ray excess is overestimated by a factor of ~2--3 in the best simulations. The scenario of tangential approach proved to be the one that best matches the properties of the profiles of X-ray surface brightness. The scenarios of radial approach of vertical approach are also marginally compatible., Comment: 15 pages, 11 figures, accepted for publication in ApJ

We present Gemini Multi-Object Spectrograph (GMOS) spectroscopic observations of 95 galaxies from the Arp & Madore (1987) catalogue of peculiar galaxies. These galaxies have been selected because they appear to be in pairs and small groups. These observations have allowed us to confirm that 60 galaxies are indeed interacting systems. For the confirmed interacting sample, we have built a matched control sample of isolated galaxies. We present an analysis of the stellar populations and nuclear activity in the interacting galaxies and compare them with the isolated galaxies. We find a median light (mass) fraction of 55% (10%) in the interacting galaxies coming from stellar populations younger than 2 Gyr and 28% (3%) in the case of the isolated galaxies. More than half of the interacting galaxies are dominated by this young stellar population, while the isolated ones have most of their light coming from older stellar populations. We used a combination of diagnostic diagrams (BPTs and WHAN) to classify the main ionization mechanisms of the gas. The interacting galaxies in our sample consistently show a higher fraction of active galactic nuclei (AGN) relative to the control sample, which ranges between 1.6 and 4 depending on the combination of diagnostic diagrams employed to classify the galaxies and the number of galaxies considered. Our study provides further observational evidence that interactions drive star formation and nuclear activity in galaxies and can have a significant impact on galaxy evolution., Comment: 42 pages, 17 figures and 15 tables. Accepted for publication in A&A

Low-rank adapters (LoRA) and their variants are popular parameter-efficient fine-tuning (PEFT) techniques that closely match full model fine-tune performance while requiring only a small number of additional parameters. These additional LoRA parameters are specific to the base model being adapted. When the base model needs to be deprecated and replaced with a new one, all the associated LoRA modules need to be re-trained. Such re-training requires access to the data used to train the LoRA for the original base model. This is especially problematic for commercial cloud applications where the LoRA modules and the base models are hosted by service providers who may not be allowed to host proprietary client task data. To address this challenge, we propose $\textit{Trans-LoRA}$ -- a novel method for lossless, nearly data-free transfer of LoRAs across base models. Our approach relies on synthetic data to transfer LoRA modules. Using large language models, we design a synthetic data generator to approximate the data-generating process of the $\textit{observed}$ task data subset. Training on the resulting synthetic dataset transfers LoRA modules to new models. We show the effectiveness of our approach using both LLama and Gemma model families. Our approach achieves lossless (mostly improved) LoRA transfer between models within and across different base model families, and even between different PEFT methods, on a wide variety of tasks.

We investigate geometrically thick disks around Kerr black holes in a swirling universe. This stationary and axisymmetric spacetime is composed of a Kerr black hole, which is immersed in a swirling background. The swirling background is characterized by an odd $\mathcal{Z}_2$ symmetry, where the northern and southern hemispheres are rotating in opposite directions. The Kerr and swirling rotations can lead to the emergence of complex spin-spin interactions, which heavily influence the spacetime properties. In order to study this influence, we analyze prograde as well as retrograde circular orbits and geometrically thick disks for different spacetime solutions, which are classified by their Kerr parameter $a$ and the swirling parameter $j$. We find stabilizing effects on prograde circular orbits and destabilizing effects on retrograde circular orbits, which originate from the spin-spin interaction and depend mainly on the Kerr rotation. Furthermore, the background rotation leads to an emergence of static orbits and the symmetry breaking regarding the equatorial plane causes a vertical distribution of the circular orbits and thick torus solutions. Due to the destabilizing effect of the swirling rotation, an outer marginally stable orbit appears, which heavily downsizes the parameter space for disk solutions. Dependent on the Keplerian specific angular momentum distribution of the spacetime and on the disk momentum $\ell_0$, different types of disk are possible which we classify and which differ in terms of the complex disk dynamics that could arise due to the presence of an inner and outer cusp or just an outer cusp.

The Lounesto classification is a well-established scheme for categorizing spinors based on their physical content, which are determined by their associated bilinear forms. It consists of six disjoint classes encompassing the known spinors within the context of the standard model of high-energy physics. However, advancements in theories beyond the standard model have opened the door to potential new spinorial adjoint structures, leading to new unforeseen classess. These developments indicate the potential for extending the standard Lounesto classification. In this paper, we explore all possible subclasses that could extend the Lounesto scheme. We highlight the most relevant subclasses by imposing constraints to their corresponding dual structures, thus broadening our understanding of spinor and its applications in theoretical physics., Comment: 7 pages; 4 tables

Near-infrared long-slit spectroscopy has been used to study the stellar population (SP) of the low luminosity active galactic nuclei (AGN) and matched analogues (LLAMA) sample. To perform the SP fits we have employed the X-shooter simple stellar population models together with the \st\ code. Our main conclusions are: The star formation history of the AGNs is very complex, presenting many episodes of star formation during their lifetimes. In general, AGN hosts have higher fractions of intermediate-age SP (light-weighted mean ages, $_L\lesssim$ 4.5 Gyr) when compared with their analogues ($_L\lesssim$ 8.0 Gyr). AGN are more affected by reddening and require significant fractions of featureless continuum and hot dust components. The ratio between the AGN radiated energy and the gravitational potential energy of the molecular gas ($E_{Rad}$/$E_{PG}$) for the AGN is compared with the \maL\ and a possible anti-correlation is observed. This suggests that the AGN is affecting the star formation in these galaxies, in the sense that more energetic AGN (log$(E_{Rad}$/$E_{PG}) \gtrsim 3$) tend to host nuclear younger SP ($ _L \lesssim$4Gyr). We found that the recent ($t<$2~Gyr) returned (recycled) stellar mass is higher in AGN than in the controls. We also provide evidence that the mass loss of stars would be enough to feed the AGN, thus providing observational constraints for models that predict that AGN feeding is partially due to the recycled gas from dying stars., Comment: Accepted for publication in MNRAS

Our previous model (NEOMOD2) for the orbital and absolute magnitude distribution of Near Earth Objects (NEOs) was calibrated on the Catalina Sky Survey observations between 2013 and 2022. Here we extend NEOMOD2 to include visible albedo information from the Wide-Field Infrared Survey Explorer. The debiased albedo distribution of NEOs can be approximated by the sum of two Rayleigh distributions with the scale parameters p_V,dark=0.03 and p_V,bright=0.17. We find evidence for smaller NEOs having (on average) higher albedos than larger NEOs; this is likely a consequence of the size-dependent sampling of different main belt sources. These inferences and the absolute magnitude distribution from NEOMOD2 are used to construct the debiased size distribution of NEOs. We estimate 830+/-60 NEOs with diameters D>1 km and 20,000+/-2,000 NEOs with D>140 m. The new model, NEOMOD3, is available via the NEOMOD Simulator -- an easy-to-operate code that can be used to generate user-defined samples (orbits, sizes and albedos) from the model., Comment: Icarus, in press

Ultra-high-energy cosmic rays, accelerated hadrons that can exceed energies of $10^{20}$ eV, are the highest-energy particles ever observed. While the sources producing UHECRs are still unknown, the Pierre Auger Observatory has detected a large-scale dipole anisotropy in the arrival directions of cosmic rays above 8 EeV. In this work, we explore whether resolved gamma-ray sources can reproduce the Auger dipole. We use various Fermi Large Area Telescope catalogs as sources of cosmic rays in CRPropa simulations. We find that in all cases, the simulated dipole has an amplitude significantly larger than that measured by Auger, even when considering large extragalactic magnetic field strengths and optimistic source weighting schemes. Our result implies that the resolved gamma-ray sources are insufficient to account for the population of sources producing the highest-energy cosmic rays, and there must exist a population of UHECR sources that lack gamma-ray emission or are unresolved by the current-generation gamma-ray telescopes., Comment: accepted for publication in ApJ-Letters

To enhance language models' cultural awareness, we design a generalizable pipeline to construct cultural knowledge bases from different online communities on a massive scale. With the pipeline, we construct CultureBank, a knowledge base built upon users' self-narratives with 12K cultural descriptors sourced from TikTok and 11K from Reddit. Unlike previous cultural knowledge resources, CultureBank contains diverse views on cultural descriptors to allow flexible interpretation of cultural knowledge, and contextualized cultural scenarios to help grounded evaluation. With CultureBank, we evaluate different LLMs' cultural awareness, and identify areas for improvement. We also fine-tune a language model on CultureBank: experiments show that it achieves better performances on two downstream cultural tasks in a zero-shot setting. Finally, we offer recommendations based on our findings for future culturally aware language technologies. The project page is https://culturebank.github.io . The code and model is at https://github.com/SALT-NLP/CultureBank . The released CultureBank dataset is at https://huggingface.co/datasets/SALT-NLP/CultureBank ., Comment: 32 pages, 7 figures, preprint

Most of the observed galaxies cannot be resolved into individual stars and are studied through their integrated spectrum using simple stellar populations (SSPs) models, with stellar libraries being a key ingredient in building them. Spectroscopic observations are increasingly being directed towards the near-infrared (NIR), where much is yet to be explored. SSPs in the NIR are still limited, and there are inconsistencies between different sets of models. One of the ways to minimize this problem is to have reliable NIR stellar libraries. The main goal of this work is to present SMARTY (mileS Moderate resolution neAr-infRared sTellar librarY) a ~0.9-2.4$\mu$m stellar spectral library composed of 31 stars observed with the Gemini Near-IR Spectrograph (GNIRS) at the 8.1m Gemini North telescope and make it available to the community. The stars were chosen from the SMARTY library, for which the atmospheric parameters are reliable (and well tested), to populate different regions of the Hertzsprung-Russell (HR) diagram. Furthermore, five of these stars have NIR spectra available that we use to assess the quality of SMARTY. The remaining 26 stars are presented for the first time in the NIR. We compared the observed SMARTY spectra with synthetic and interpolated spectra, finding a mean difference of ~20% in the equivalent widths and ~1% in the overall continuum shape in both sets of comparisons. We computed the spectrophotometric broadband magnitudes and colours and compared them with the 2MASS ones, resulting in mean differences up to 0.07 and 0.10mag in magnitudes and colours, respectively. In general, a small difference was noted between the SMARTY spectra corrected using the continuum from the interpolated and the theoretical stars.

We study the ionised gas kinematics of 293 Active Galactic Nuclei (AGN) hosts as compared to that of 485 control galaxies from the MaNGA-SDSS survey using measurements of the [OIII]$\lambda$5007\AA emission-line profiles, presenting flux, velocity and W$_{80}$ maps. In 45% of the AGN, a broad component was needed to fit the line profiles wings within the inner few kpc, that we have identified with an outflow. But in most AGN, the profiles are broader than that of their controls over a much more extended region, identified as the "kinematically disturbed regions" (KDRs). We find a positive correlation between the mean $\langle$W$_{80}\rangle$ and L[OIII], supporting that the KDR is due to heating and turbulence of the ISM by outflows and radiation from the AGN. The extent R$_{KDR}$ reaches up to 24kpc, with a mean ratio to that of the ENLR of 57%. We estimate ionised gas mass flow rates ($\dot{M}_{\rm out}$) and kinetic powers ($\dot{E}_{\rm out}$) both from the AGN broad components and from the W$_{80}$ values, that can be obtained for the whole AGN sample. We find values for $\dot{M}_{\rm out}$ and $\dot{E}_{\rm out}$ that correlate with the AGN luminosity $L_{bol}$, populating the low luminosity end of these known correlations. The mean coupling efficiency between $\dot{E}_{\rm out}$ and AGN luminosity is $\approx$ 0.02% from the W$_{80}$ values and lower from the broad component. But the large extent of the KDR shows that even low-luminosity AGN can impact the host galaxy along several kpc in a "maintenance mode" feedback.

Foundation Models (FMs) have become the hallmark of modern AI, however, these models are trained on massive data, leading to financially expensive training. Updating FMs as new data becomes available is important, however, can lead to `catastrophic forgetting', where models underperform on tasks related to data sub-populations observed too long ago. This continual learning (CL) phenomenon has been extensively studied, but primarily in a setting where only a small amount of past data can be stored. We advocate for the paradigm where memory is abundant, allowing us to keep all previous data, but computational resources are limited. In this setting, traditional replay-based CL approaches are outperformed by a simple baseline which replays past data selected uniformly at random, indicating that this setting necessitates a new approach. We address this by introducing a framework of adaptive memory replay for continual learning, where sampling of past data is phrased as a multi-armed bandit problem. We utilize Bolzmann sampling to derive a method which dynamically selects past data for training conditioned on the current task, assuming full data access and emphasizing training efficiency. Through extensive evaluations on both vision and language pre-training tasks, we demonstrate the effectiveness of our approach, which maintains high performance while reducing forgetting by up to 10% at no training efficiency cost., Comment: CVPR-W 2024 (Spotlight)