Search

Complex organic molecules (COMs) are widely detected in protostellar and protoplanetary systems, where they are thought to have been inherited in large part from earlier evolutionary phases. The chemistry of COMs in these earlier phases, namely starless and prestellar cores, remains poorly understood, as models often struggle to reproduce the observed gas-phase abundances of these species. We simulate the formation of a molecular cloud, and the cores within it, out of the diffuse interstellar medium, and follow the chemical evolution of the cloud material starting from purely-atomic initial conditions. We find that the formation of both gas- and ice-phase COMs precedes the formation of cores as distinct objects, beginning at gas densities of a few $10^3 \,{\rm cm}^{-3}$. Much of this COM-enriched material remains at these relatively modest densities for several Myr, which may provide a reservoir for accretion onto planet-forming discs in later evolutionary stages. We suggest that models of core and disc chemistry should not ignore the complex dynamical evolution which precedes these structures, even when studying supposedly late-forming molecules such as CH$_3$OH and CH$_3$CN., Comment: 10 pages, 8 figures. MNRAS accepted

Current language models (LMs) use a fixed, static subword tokenizer. This default choice typically results in degraded efficiency and language capabilities, especially in languages other than English. To address this issue, we challenge the static design and propose retrofitting LMs with dynamic tokenization: a way to dynamically decide on token boundaries based on the input text via a subword-merging algorithm inspired by byte-pair encoding. We merge frequent subword sequences in a batch, then apply a pre-trained embedding-prediction hypernetwork to compute the token embeddings on-the-fly. For encoder-style models (e.g., XLM-R), this on average reduces token sequence lengths by >20% across 14 languages while degrading performance by less than 2%. The same method applied to pre-filling and scoring in decoder-style models (e.g., Mistral-7B; evaluated on English) results in minimal performance degradation at up to 6% reduction in sequence length. Overall, we find that dynamic tokenization can mitigate the limitations of static tokenization by substantially improving inference speed and promoting fairness across languages, enabling more equitable and adaptable LMs.

In an era increasingly dominated by digital platforms, the spread of misinformation poses a significant challenge, highlighting the need for solutions capable of assessing information veracity. Our research contributes to the field of Explainable Artificial Antelligence (XAI) by developing transformer-based fact-checking models that contextualise and justify their decisions by generating human-accessible explanations. Importantly, we also develop models for automatic evaluation of explanations for fact-checking verdicts across different dimensions such as \texttt{(self)-contradiction}, \texttt{hallucination}, \texttt{convincingness} and \texttt{overall quality}. By introducing human-centred evaluation methods and developing specialised datasets, we emphasise the need for aligning Artificial Intelligence (AI)-generated explanations with human judgements. This approach not only advances theoretical knowledge in XAI but also holds practical implications by enhancing the transparency, reliability and users' trust in AI-driven fact-checking systems. Furthermore, the development of our metric learning models is a first step towards potentially increasing efficiency and reducing reliance on extensive manual assessment. Based on experimental results, our best performing generative model \textsc{ROUGE-1} score of 47.77, demonstrating superior performance in generating fact-checking explanations, particularly when provided with high-quality evidence. Additionally, the best performing metric learning model showed a moderately strong correlation with human judgements on objective dimensions such as \texttt{(self)-contradiction and \texttt{hallucination}, achieving a Matthews Correlation Coefficient (MCC) of around 0.7.}, Comment: Forthcoming in Engineering Applications of Artificial Intelligence

We study strongly isochronous Hamiltonians that generate periodic time evolution with the same basic period for a dense set of initial values. We explain that all such Hamiltonians are maximally superintegrable, and show that if the system is subjected to Hamiltonian reduction based on a compact symmetry group and certain conditions are met, then the reduced Hamiltonian is strongly isochronous with the original basic period. We utilize these simple observations for demonstrating the maximal superintegrability of rational spin Calogero--Moser type models in confining harmonic potential., Comment: 19 pages, v2: corrected 3 minor sign errors

We present synthetic line observations of a simulated molecular cloud, utilising a self-consistent treatment of the dynamics and time-dependent chemical evolution. We investigate line emission from the three most common CO isotopologues ($^{12}$CO, $^{13}$CO, C$^{18}$O) and six supposed tracers of dense gas (NH$_3$, HCN, N$_2$H$^+$, HCO$^+$, CS, HNC). Our simulation produces a range of line intensities consistent with that observed in real molecular clouds. The HCN-to-CO intensity ratio is relatively invariant with column density, making HCN (and chemically-similar species such as CS) a poor tracer of high-density material in the cloud. The ratio of N$_2$H$^+$ to HCN or CO, on the other hand, is highly selective of regions with densities above $10^{22} \, {\rm cm^{-2}}$, and the N$_2$H$^+$ line is a very good tracer of the dynamics of high volume density ($>10^4 \, {\rm cm^{-3}}$) material. Focusing on cores formed within the simulated cloud, we find good agreement with the line intensities of an observational sample of prestellar cores, including reproducing observed CS line intensities with an undepleted elemental abundance of sulphur. However, agreement between cores formed in the simulation, and models of isolated cores which have otherwise-comparable properties, is poor. The formation from and interaction with the large-scale environment has a significant impact on the line emission properties of the cores, making isolated models unsuitable for interpreting observational data., Comment: 14 pages, 14 figures. MNRAS accepted

The Large Hadron Collider will soon undergo an upgrade to increase its luminosity by a factor of ~10 [1]. A crucial part of this upgrade will be replacement of the NbTi focusing magnets with Nb3Sn magnets that achieve a ~50% increase in the field strength. This will be the first ever large-scale implementation of Nb3Sn magnets in a particle accelerator. The High-Luminosity LHC Upgrade, HL-LHC is a CERN project with a world-wide collaboration. It is under construction and utilizes Nb3Sn Magnets (named MQXF) as key ingredients to increase tenfold the integrated luminosity delivered to the CMS and ATLAS experiments in the next decade. The HL-LHC AUP is the US effort to contribute approximately 50% of the low-beta focusing magnets and crab cavities for the HL-LHC. This paper will present the program to fabricate the Nb3Sn superconducting magnets. We are reporting the status of the HL-LHC AUP project present the results from horizontal tests of the first fully assembled cryo-assembly., Comment: IPAC'24 - 15th International Particle Accelerator Conference

The COVID-19 pandemic led to a large global effort to sequence SARS-CoV-2 genomes from patient samples to track viral evolution and inform public health response. Millions of SARS-CoV-2 genome sequences have been deposited in global public repositories. The Canadian COVID-19 Genomics Network (CanCOGeN - VirusSeq), a consortium tasked with coordinating expanded sequencing of SARS-CoV-2 genomes across Canada early in the pandemic, created the Canadian VirusSeq Data Portal, with associated data pipelines and procedures, to support these efforts. The goal of VirusSeq was to allow open access to Canadian SARS-CoV-2 genomic sequences and enhanced, standardized contextual data that were unavailable in other repositories and that meet FAIR standards (Findable, Accessible, Interoperable and Reusable). The Portal data submission pipeline contains data quality checking procedures and appropriate acknowledgement of data generators that encourages collaboration. Here we also highlight Duotang, a web platform that presents genomic epidemiology and modeling analyses on circulating and emerging SARS-CoV-2 variants in Canada. Duotang presents dynamic changes in variant composition of SARS-CoV-2 in Canada and by province, estimates variant growth, and displays complementary interactive visualizations, with a text overview of the current situation. The VirusSeq Data Portal and Duotang resources, alongside additional analyses and resources computed from the Portal (COVID-MVP, CoVizu), are all open-source and freely available. Together, they provide an updated picture of SARS-CoV-2 evolution to spur scientific discussions, inform public discourse, and support communication with and within public health authorities. They also serve as a framework for other jurisdictions interested in open, collaborative sequence data sharing and analyses.

Recent advances in identifying giant molecular filaments in galactic surveys allow us to study the interstellar material and its dense, potentially star forming phase on scales comparable to resolved extragalactic clouds. Two large filaments detected in the CHIMPS $^{13}$CO(3-2) survey, one in the Sagittarius-arm and one in an inter-arm region, were mapped with dense gas tracers inside a 0.06 deg$^2$ area and with a spatial resolution of around 0.4 and 0.65 pc at the distance of the targets using the IRAM 30m telescope, to investigate the environmental dependence of the dense gas fraction. The N$_2$H$^+$(1-0) transition, an excellent tracer of the dense gas, was detected in parsec-scale, elliptical clumps and with a filling factor of around 8.5% in our maps. The N$_2$H$^+$-emitting areas appear to have higher dense gas fraction (e.g. the ratio of N$_2$H$^+$ and $^{13}$CO emission) in the inter-arm than in the arm which is opposite to the behaviour found by previous studies, using dust emission rather than N$_2$H$^+$ as a tracer of dense gas. However, the arm filament is brighter in $^{13}$CO and the infrared emission of dust, and the dense gas fraction determined as above is governed by the $^{13}$CO brightness. We caution that measurements regarding the distribution and fraction of dense gas on these scales may be influenced by many scale- and environment-dependent factors, as well as the chemistry and excitation of the particular tracers, then consider several scenarios that can reproduce the observed effect., Comment: 18 pages, 9 figures, accepted by MNRAS

The High Luminosity upgrade of the Large Hadron Collider (HL-LHC) at CERN will include eight cryo-assemblies that are expected to be fabricated and delivered to CERN by the US HL-LHC Accelerator Upgrade Project (AUP) as part of the U.S. contributions to the HL-LHC. These cryostat assemblies are the quadrupole magnetic components of the HL-LHC Q1 and Q3 inner triplet optical elements in front of the two interaction points. Each cryo-assembly consists of two 4.2 m long Nb3Sn quadrupole magnets with aperture 150 mm and operating gradient 132.6 T/m. The first pre-series cryo-assembly has been fabricated and successfully tested at the horizontal test facility at Fermi National Accelerator Laboratory. In this manuscript we report the quench test results of the LQXFA/B-01 cryo-assembly. The primary objective of the horizontal test is full cryo-assembly qualification and validation of the performance requirements., Comment: MT28 International Conference on Magnet Technology, Accepted Version

The high-luminosity LHC upgrade requires stronger than LHC low-beta quadrupole magnets to reach the luminosity goals of the project. The project is well advanced and HL-LHC quadrupole magnets are currently being commissioned in US Labs (MQXFA magnets) and CERN (MQXFB magnets). Those are the first Nb3Sn magnets to be used in any large particle accelerator. At development stages, many Nb3Sn accelerator sub-scale models showed relatively slow training and MQXFA magnets were projected to have low tens of quenches before reaching operational field. Recently it was shown that dedicated capacitor-based devices can affect Nb3Sn magnet training, and it was suggested that CLIQ, a capacitor-based device intended for quench protection, can do too. The present paper investigates effects on training likely induced by CLIQ, using the base fact that only half the coils in a quadrupole experience upward current modulation at quench because of capacitor discharge. The study encompasses all MQXFA production magnets trained at BNL to date. No other high-statistics data from identical magnets (series) with CLIQ protection exist so far. Implications and opportunities stemming from data analysis are discussed and conclusions drawn., Comment: Accepted version

Some generalizations of spin Sutherland models descend from `master integrable systems' living on Heisenberg doubles of compact semisimple Lie groups. The master systems represent Poisson--Lie counterparts of the systems of free motion modeled on the respective cotangent bundles and their reduction relies on taking quotient with respect to a suitable conjugation action of the compact Lie group. We present an enhanced exposition of the reductions and prove rigorously for the first time that the reduced systems possess the property of degenerate integrability on the dense open subset of the Poisson quotient space corresponding to the principal orbit type for the pertinent group action. After restriction to a smaller dense open subset, degenerate integrability on the generic symplectic leaves is demonstrated as well. The paper also contains a novel description of the reduced Poisson structure and a careful elaboration of the scaling limit whereby our reduced systems turn into the spin Sutherland models., Comment: 32 pages, v2: slightly modified the abstract and corrected small typos

The symmetries of a spinning particle in the field of a self-dual monopole are studied from the viewpoint of supersymmetric quantum mechanics., Comment: Originally appeared in: Symmetries in Science III, Proc. Schloss Hofen Meeting, Gruber and Iachello (eds.), pp. 525-529. Plenum (1989). Available in inspire (code {Feher:1988ym}). Typos corrected and reference list updated. 7 pages

MQXFA production series quadrupole magnets are being built for the Hi-Lumi (HL) LHC upgrade by the US Accelerator Upgrade Project (US-HL-LHC AUP). These magnets are being placed in pairs, as a cold mass, within cryostats at Fermilab, and are being tested to assess alignment and magnetic performance at Fermilab's horizontal test stand facility. The ~10 m - long assembly must meet stringent specifications for quadrupole strength and harmonic field integrals determination, magnetic axis position, and for magnet variations in positioning and local field profile. This paper describes the results of the magnetic and alignment measurements which characterize the first LQXFA/B assembly., Comment: 28th International Conference on Magnet Technology (MT-28)

We give a new method to calculate the universal cohomology classes of coincident root loci. We show a polynomial behavior of them and apply this result to prove that generalized Pl\"ucker formulas are polynomials in the degree, just as the classical Pl\"ucker formulas counting the bitangents and flexes of a degree $d$ generic plane curve. We establish an upper bound for the degrees of these polynomials, and we calculate the leading terms of those whose degrees reach this upper bound. We believe that the paper is understandable without detailed knowledge of equivariant cohomology. It may serve as a demonstration of the use of equivariant cohomology in enumerative geometry through the examples of coincident root strata. We also explain how the equivariant method can be "translated" into the traditional non-equivariant method of resolutions., Comment: 29 pages; abstract made more precise, reference added

We propose a geometrically motivated mathematical model which reveals the key features of coastal and fluvial fragment shape evolution from the earliest stages of the abrasion. Our \textit{collisional polygon model} governs the evolution through an ordinary differential equation (ODE) that determines the rounding rate of initially sharp corners in the function of the size reduction. As an approximation, the basic structure of our model adopts the concept of Bloore's partial differential equation (PDE) in terms of the curvature dependent local collisional frequency. We tested our model under various conditions and made comparisons with the predictions of Bloore's PDE. Moreover, we applied the model to discover and quantify the mathematical conditions corresponding to typical and special shape evolution. By further extending our model to investigate the self-dual and mixed cases, we outline a possible explanation of the long-term preservation of initial pebble shape characteristics., Comment: 9 pages, 7 figures

Star formation activity in molecular clouds is often found to be correlated with the amount of material above a column density threshold of $\sim 10^{22} \, {\rm cm^{-2}}$. Attempts to connect this column density threshold to a ${\it volume}$ density above which star formation can occur are limited by the fact that the volume density of gas is difficult to reliably measure from observations. We post-process hydrodynamical simulations of molecular clouds with a time-dependent chemical network, and investigate the connection between commonly-observed molecular species and star formation activity. We find that many molecules widely assumed to specifically trace the dense, star-forming component of molecular clouds (e.g. HCN, HCO$^+$, CS) actually also exist in substantial quantities in material only transiently enhanced in density, which will eventually return to a more diffuse state without forming any stars. By contrast, N$_2$H$^+$ only exists in detectable quantities above a volume density of $10^4 \, {\rm cm^{-3}}$, the point at which CO, which reacts destructively with N$_2$H$^+$, begins to deplete out of the gas phase onto grain surfaces. This density threshold for detectable quantities of N$_2$H$^+$ corresponds very closely to the volume density at which gas becomes irreversibly gravitationally bound in the simulations: the material traced by N$_2$H$^+$ never reverts to lower densities, and quiescent regions of molecular clouds with visible N$_2$H$^+$ emission are destined to eventually form stars. The N$_2$H$^+$ line intensity is likely to directly correlate with the star formation rate averaged over timescales of around a Myr., Comment: 10 pages, 10 figures. MNRAS accepted

The reduction of the `master system' of free motion on the cotangent bundle $T^*G$ of a compact, connected and simply connected, semisimple Lie group is considered using the conjugation action of $G$. It is proved that the restriction of the reduced system to the smooth component of the quotient space $T^*G/G$, given by the principal orbit type, inherits the degenerate integrability of the master system. The proof can be generalized easily to other interesting examples of Hamiltonian reduction., Comment: 14 pages, v2: corrected equation (4.54), simplified the proof of Lemma A.1 and removed typos, v3: updated references and removed newly noticed typos (equations unchanged)

Approximate Nearest Neighbor Search (ANNS) plays a critical role in various disciplines spanning data mining and artificial intelligence, from information retrieval and computer vision to natural language processing and recommender systems. Data volumes have soared in recent years and the computational cost of an exhaustive exact nearest neighbor search is often prohibitive, necessitating the adoption of approximate techniques. The balanced performance and recall of graph-based approaches have more recently garnered significant attention in ANNS algorithms, however, only a few studies have explored harnessing the power of GPUs and multi-core processors despite the widespread use of massively parallel and general-purpose computing. To bridge this gap, we introduce a novel parallel computing hardware-based proximity graph and search algorithm. By leveraging the high-performance capabilities of modern hardware, our approach achieves remarkable efficiency gains. In particular, our method surpasses existing CPU and GPU-based methods in constructing the proximity graph, demonstrating higher throughput in both large- and small-batch searches while maintaining compatible accuracy. In graph construction time, our method, CAGRA, is 2.2~27x faster than HNSW, which is one of the CPU SOTA implementations. In large-batch query throughput in the 90% to 95% recall range, our method is 33~77x faster than HNSW, and is 3.8~8.8x faster than the SOTA implementations for GPU. For a single query, our method is 3.4~53x faster than HNSW at 95% recall., Comment: Accepted to ICDE 2024

Materials with complex inner structure can be challenging to characterize in an effective way. First, it is difficult to determine the representative volume for such a heterogeneous material. Second, the effective material parameters significantly depend on the structure, and thus on the interface properties. In the present paper, we focus on composite metal foams, and we attempt to determine the effective thermal parameters. We use the heat pulse experiment to study its transient thermal response. We observed deviation from Fourier's law, and thus we propose an evaluation procedure for such experimental data using the Guyer--Krumhansl heat equation. Furthermore, we studied the effective parameters, the specific heat, mass density, thermal conductivity and thermal diffusivity, and we propose a method for deducing a more reliable thermal diffusivity and thermal conductivity based on the transient temperature history.

Data compression algorithms typically rely on identifying repeated sequences of symbols from the original data to provide a compact representation of the same information, while maintaining the ability to recover the original data from the compressed sequence. Using data transformations prior to the compression process has the potential to enhance the compression capabilities, being lossless as long as the transformation is invertible. Floating point data presents unique challenges to generate invertible transformations with high compression potential. This paper identifies key conditions for basic operations of floating point data that guarantee lossless transformations. Then, we show four methods that make use of these observations to deliver lossless compression of real datasets, where we improve compression rates up to 40 %.

Much of what we know about molecular clouds, and by extension star formation, comes from molecular line observations. Interpreting these correctly requires knowledge of the underlying molecular abundances. Simulations of molecular clouds typically only model species that are important for the gas thermodynamics, which tend to be poor tracers of the denser material where stars form. We construct a framework for post-processing these simulations with a full time-dependent chemical network, allowing us to model the behaviour of observationally-important species not present in the reduced network used for the thermodynamics. We use this to investigate the chemical evolution of molecular gas under realistic physical conditions. We find that molecules can be divided into those which reach peak abundances at moderate densities ($10^3 \, {\rm cm^{-3}}$) and decline sharply thereafter (such as CO and HCN), and those which peak at higher densities and then remain roughly constant (e.g. NH$_3$, N$_2$H$^+$). Evolving the chemistry with physical properties held constant at their final values results in a significant overestimation of gas-phase abundances for all molecules, and does not capture the drastic variations in abundance caused by different evolutionary histories. The dynamical evolution of molecular gas cannot be neglected when modelling its chemistry., Comment: 14 pages, 13 figures. MNRAS accepted

Graphite has been intensively studied, yet its electron spins dynamics remains an unresolved problem even 70 years after the first experiments. The central quantities, the longitudinal ($T_1$) and transverse ($T_2$) relaxation times were postulated to be equal, mirroring standard metals, but $T_1$ has never been measured for graphite. Here, based on a detailed band structure calculation including spin-orbit coupling, we predict an unexpected behavior of the relaxation times. We find, based on saturation ESR measurements, that $T_1$ is markedly different from $T_2$. Spins injected with perpendicular polarization with respect to the graphene plane have an extraordinarily long lifetime of $100$ ns at room temperature. This is ten times more than in the best graphene samples. The spin diffusion length across graphite planes is thus expected to be ultralong, on the scale of $\sim 70~\mu$m, suggesting that thin films of graphite -- or multilayer AB graphene stacks -- can be excellent platforms for spintronics applications compatible with 2D van der Waals technologies. Finally, we provide a qualitative account of the observed spin relaxation based on the anisotropic spin admixture of the Bloch states in graphite obtained from density functional theory calculations.

We give families of knots and links with pairs of Seifert surfaces that are topologically non-isotopic in $D^4$. This generalizes the main example of Hayden-Kim-Miller-Park-Sundberg and the proof is similarly based on the double branched cover and the Seifert form. Moreover, using Conway's theory of topographs, we implement an algorithm that decides whether two integral quadratic forms in two variables are isomorphic over $\mathbb Z$., Comment: 17 pages, 10 figures. In v2, added Theorems 3-5 and added Section 2

The future High Luminosity upgrade of the Large Hadron Collider (HL-LHC) at CERN will include the low-beta inner triplets (Q1, Q2a/b, Q3) for two LHC insertion regions. The Q1, Q3 components consist of eight 10 m-long LMQXFA cryo-assemblies fabricated by the HL-LHC Accelerator Upgrade Project. Each LMQXFA Cold mass contains two Nb3Sn magnets connected in series. A stainless-steel shell is welded around the two magnets before the insertion into the cryostat. There is a limit on how much coil preload increase induced by the shell welding is allowed. Distributed Rayleigh backscattering fiber optics sensors were used for the first time to obtain a strain map over a wide area of a Nb3Sn magnet cold mass shell. Data were collected during welding of the first LMQXFA cold mass and the results confirm that the increase of the coil pole azimuthal pre-stress after welding do not exceed requirements.

The number of IoT devices is expected to continue its dramatic growth in the coming years and, with it, a growth in the amount of data to be transmitted, processed and stored. Compression techniques that support analytics directly on the compressed data could pave the way for systems to scale efficiently to these growing demands. This paper proposes two novel methods for preprocessing a stream of floating point data to improve the compression capabilities of various IoT data compressors. In particular, these techniques are shown to be helpful with recent compressors that allow for random access and analytics while maintaining good compression. Our techniques improve compression with reductions up to 80% when allowing for at most 1% of recovery error.

We construct a master dynamical system on a $\operatorname{U}(n)$ quasi-Poisson manifold, $\mathcal{M}_d$, built from the double $\operatorname{U}(n) \times \operatorname{U}(n)$ and $d\geq 2$ open balls in $\mathbb{C}^n$, whose quasi-Poisson structures are obtained from $T^* \mathbb{R}^n$ by exponentiation. A pencil of quasi-Poisson bivectors $P_{\underline{z}}$ is defined on $\mathcal{M}_d$ that depends on $d(d-1)/2$ arbitrary real parameters and gives rise to pairwise compatible Poisson brackets on the $\operatorname{U}(n)$-invariant functions. The master system on $\mathcal{M}_d$ is a quasi-Poisson analogue of the degenerate integrable system of free motion on the extended cotangent bundle $T^*\!\operatorname{U}(n) \times \mathbb{C}^{n\times d}$. Its commuting Hamiltonians are pullbacks of the class functions on one of the $\operatorname{U}(n)$ factors. We prove that the master system descends to a degenerate integrable system on a dense open subset of the smooth component of the quotient space $\mathcal{M}_d/\operatorname{U}(n)$ associated with the principal orbit type. Any reduced Hamiltonian arising from a class function generates the same flow via any of the compatible Poisson structures stemming from the bivectors $P_{\underline{z}}$. The restrictions of the reduced system on minimal symplectic leaves parameterized by generic elements of the center of $\operatorname{U}(n)$ provide a new real form of the complex, trigonometric spin Ruijsenaars-Schneider model of Krichever and Zabrodin. This generalizes the derivation of the compactified trigonometric RS model found previously in the $d=1$ case., Comment: 56 pages, 2 figures, v2: a typo is removed from equation (4.25)

Superconducting magnet training is one of the accelerator related issues attracting attention due to significant operational costs and time budget associated to it. It is especially worrisome that magnets based on the next-generation Nb3Sn technology are affected by long training. While various efforts are underway to better understand and resolve the problem a parallel path could also be investigated, a path bypassing the issue. Following the concept of fast induced over-current during magnet powering, FNAL has developed an upgradable capacitor-based device to discharge through a superconducting magnet at quench detection or operator chosen time. The 0.4 F/1 kV device has been tested on a 1-m-long dipole-coil in a mirror magnet configuration and conclusive results on magnet training elimination have been observed. In this paper we discuss the main characteristics of the device, compare simulated response and actual performance, elaborate on test drivers and outcomes. Next steps and perspectives for future use are debated.

By the end of October 2022, the US HL-LHC Accelerator Upgrade Project (AUP) had completed fabrication of ten MQXFA magnets and tested eight of them. The MQXFA magnets are the low beta quadrupole magnets to be used in the Q1 and Q3 Inner Triplet elements of the High Luminosity LHC. This AUP effort is shared by BNL, Fermilab, and LBNL, with strand verification tests at NHMFL. An important step of the AUP QA plan is the testing of MQXFA magnets in a vertical cryostat at BNL. The acceptance criteria that could be tested at BNL were all met by the first four production magnets (MQXFA03-MQXFA06). Subsequently, two magnets (MQXFA07 and MQXFA08) did not meet some criteria and were disassembled. Lessons learned during the disassembly of MQXFA07 caused a revision to the assembly specifications that were used for MQXFA10 and subsequent magnets. In this paper, we present a summary of: 1) the fabrication and test data of all the MQXFA magnets; 2) the analysis of MQXFA07/A08 test results with characterization of the limiting mechanism; 3) the outcome of the investigation, including the lessons learned during MQXFA07 disassembly; and 4) the finite element analysis correlating observations with test performance.

FU Orionis-type objects (FUors) are low-mass pre-main-sequence objects which go through a short-lived phase (~100 years) of increased mass accretion rate (from 10^-8 to 10^-4 M_sun yr^-1). These eruptive young stars are in the early stages of stellar evolution and, thus, still deeply embedded in a massive envelope that feeds material to the circumstellar disk that is then accreted onto the star. Some FUors drive molecular outflows, i.e. low-velocity wide-angle magneto-hydrodynamical winds, that inject energy and momentum back to the surrounding envelopes, and help clear the material surrounding the young star. Here we present a 12CO (3--2), 13CO (3--2) and 12CO (4--3) survey of 20 FUor-type eruptive young stars observed with APEX. We use our 13CO (3--2) observations to measure the masses of the envelopes surrounding each FUor and find an agreement with the FUor evolutionary trend found from the 10um silicate feature. We find outflows in 11 FUors, calculate their masses and other kinematic properties, and compare these with those of outflows found around quiescent young stellar objects gathered from the literature. This comparison indicates that outflows in FUors are more massive than outflows in quiescent sources, and that FUor outflows have a higher ratio outflow mass with respect to the envelope than the quiescent sample, indicating that the eruptive young stars have lower star-forming efficiencies. Finally, we found that the outflow forces in FUors are similar to those of quiescent young stellar objects, indicating that their accretion histories are similar or that the FUor outflows have lower velocities., Comment: Accepted for publication in ApJ. 29 pages, 8 tables, 10 figures, 2 figuresets