This paper presents the winning submission of the RaaVa team to the AmericasNLP 2025 Shared Task 3 on Automatic Evaluation Metrics for Machine Translation (MT) into Indigenous Languages of America, where our system ranked first overall based on average Pearson correlation with the human annotations. We introduce Feature-Union Scorer (FUSE) for Evaluation, FUSE integrates Ridge regression and Gradient Boosting to model translation quality. In addition to FUSE, we explore five alternative approaches leveraging different combinations of linguistic similarity features and learning paradigms. FUSE Score highlights the effectiveness of combining lexical, phonetic, semantic, and fuzzy token similarity with learning-based modeling to improve MT evaluation for morphologically rich and low-resource languages. MT into Indigenous languages poses unique challenges due to polysynthesis, complex morphology, and non-standardized orthography. Conventional automatic metrics such as BLEU, TER, and ChrF often fail to capture deeper aspects like semantic adequacy and fluency. Our proposed framework, formerly referred to as FUSE, incorporates multilingual sentence embeddings and phonological encodings t
Since its inception the memristive fuse has been a good example of how small numbers of memristors can be combined to obtain useful behaviours unachievable by individual devices. In this work, we link the memristive fuse concept with that of the Complementary Resistive Switch (CRS), exploit that link to experimentally demonstrate a practical memristive fuse using TiOx-based ReRAM cells and explain its basic operational principles. The fuse is stimulated by trains of identical pulses where successive pulse trains feature opposite polarities. In response, we observe a gradual (analogue) drop in resistive state followed by a gradual recovery phase regardless of input stimulus polarity; echoing traditional, binary CRS behaviour. This analogue switching property opens the possibility of operating the memristive fuse as a single-component step change detector. Moreover, we discover that the characteristics of the individual memristors used to demonstrate the memristive fuse concept in this work allow our fuse to be operated in a regime where one of the two constituent devices can be switched largely independently from the other. This property, not present in the traditional CRS, indicate
Polysilicon fuses are one time programmable memory elements which allow the calibration of integrated circuits at wafer and package level. We present a zero dimensional lumped parameter model of the programming of fuses made from a combination of tungsten silicide and polycrystalline silicon. The components of the model are an electrical model, a thermal model and a flow model. The electrical model describes the temperature and geometry dependent resistance of the fuse. The thermal model describes the heating and melting of the fuse and its surroundings. The flow model describes the disconnection of the fuse by electromigration driven flow of silica. The model generates quantitatively accurate results and reproduces trends with applied voltage and fuse size.
Fiber fuse effect can occur spontaneously and propagate along optical fibers to cause wide-spread damage; it threatens all applications involving optical fibers. This paper presents two results. First, it establishes that the initiation of fiber fuse (IFF) in silica fibers is caused by defect-induced absorption. Critical temperatures and critical optical powers for IFF are simulated for the first time using a 3D solid-state heat transfer model with heat source generated by defect-induced absorption. In this method, formation energies of the defects can be uniquely determined, which offers critical information on the chemical reasons for fiber fuse. Second, this paper offers a method to evaluate operating temperatures of fiber lasers. General analytical solutions of the operating temperatures along gain fibers are deduced. Results of 976-nm laser-diode-(LD)-pumped and 1018-nm tandem-pumped ytterbium-doped fiber (YDF) amplifiers using 10/130-μm YDFs are calculated. Potential limits caused by fiber fuse are discussed.
We present numerical studies of electrical breakdown in disordered materials using a two-dimensional thermal fuse model with heat diffusion. A conducting fuse is heated locally by a Joule heating term. Heat diffuses to neighbouring fuses by a diffusion term. When the temperature reaches a given threshold, the fuse breaks and turns into an insulator. The time dynamics is governed by the time scales related to the two terms, in the presence of quenched disorder in the conductances of the fuses. For the two limiting domains, when one time scale is much smaller than the other, we find that the global breakdown time $t_r$ follows $t_r\sim I^2$ and $t_r\sim L^2$, where $I$ is the applied current, and $L$ is the system size. However, such power law does not apply in the intermediate domain where the competition between the two terms produces a subtle behaviour.
We report the results of initial FUSE observations of molecular hydrogen (H2) in translucent clouds. These clouds have greater optical depth than any of the diffuse clouds previously observed for far-UV H2 absorption, and provide new insights into the physics and chemistry of such regions. Our initial results involve observations of HD 73882, a well-studied southern hemisphere star lying behind substantial interstellar material (E(B-V) = 0.72; A_V = 2.44). We find a total H2 column density, N(H2) = 1.2 x 10^{21} cm^{-2}, about three times larger than the values for diffuse clouds previously measured in the far-UV. The gas kinetic temperature indicated by the ratio N(J=1)/N(J=0) is 58 +/- 10 K. With the aid of ground-based data to calculate an appropriate multi-component curve of growth, we have determined column densities for all rotational levels up to J = 7. The J >= 2 states can be reasonably fitted with a rotational excitation temperature of 307 +/- 23 K. Both the kinetic and rotational temperatures are similar to those found in previous investigations of diffuse clouds. The ratios of carbonaceous molecules to hydrogen molecules are also similar to ratios in diffuse clouds,
Aerial vehicles are no longer limited to flying in open space: recent work has focused on aerial manipulation and up-close inspection. Such applications place stringent requirements on state estimation: the robot must combine state information from many sources, including onboard odometry and global positioning sensors. However, flying close to or in contact with structures is a degenerate case for many sensing modalities, and the robot's state estimation framework must intelligently choose which sensors are currently trustworthy. We evaluate a number of metrics to judge the reliability of sensing modalities in a multi-sensor fusion framework, then introduce a consensus-finding scheme that uses this metric to choose which sensors to fuse or not to fuse. Finally, we show that such a fusion framework is more robust and accurate than fusing all sensors all the time and demonstrate how such metrics can be informative in real-world experiments in indoor-outdoor flight and bridge inspection.
We report on FUSE spectra of three U Gem-type, long period, dwarf novae, UU Aql, BV Cen and CH UMa taken during their quiescence intervals. We discuss the line identifications in their spectra and attempt to characterize the source(s) of their FUV flux distribution. Archival IUE spectrum of CH UMa and BV Cen in quiescence were identified as having a matching flux level with the FUSE spectra and these were combined with each FUSE spectrum to broaden the wavelength coverage and further constrain model fits. Multi-component synthetic spectral fits from our model grids, consisting of single temperature white dwarfs, two-temperature white dwarfs, accretion disks and white dwarfs plus accretion disks, were applied to the FUSE spectra alone and to the combined FUSE + IUE spectra. We present the results of our model analyses and their implications.
I review the five years of progress by FUSE on current topics in the interstellar medium. FUSE's sensitivity and unique access to the far ultraviolet allow investigators to solve problems in all phases of the interstellar medium. I describe FUSE's contributions in four major areas: 1) the Local Interstellar Medium (LISM), 2) the hot phase (O VI), 3) the cold phase (H_2), and 4) interstellar gas abundances. I devote particular attention to the common themes of ISM phase interactions and changes with metallicity. As a whole, these results show that FUSE is the most powerful machine ever to address problems of the ISM, and that FUSE points vividly to the future of ISM studies in the Galaxy, Local Group, and beyond.
Bloom and cuckoo filters provide fast approximate set membership while using little memory. Engineers use them to avoid expensive disk and network accesses. The recently introduced xor filters can be faster and smaller than Bloom and cuckoo filters. The xor filters are within 23% of the theoretical lower bound in storage as opposed to 44% for Bloom filters. Inspired by Dietzfelbinger and Walzer, we build probabilistic filters -- called binary fuse filters -- that are within 13% of the storage lower bound -- without sacrificing query speed. As an additional benefit, the construction of the new binary fuse filters can be more than twice as fast as the construction of xor filters. By slightly sacrificing query speed, we further reduce storage to within 8% of the lower bound. We compare the performance against a wide range of competitive alternatives such as Bloom filters, blocked Bloom filters, vector quotient filters, cuckoo filters, and the recent ribbon filters. Our experiments suggest that binary fuse filters are superior to xor filters.
We propose a mean field theory for the localization of damage in a quasistatic fuse model on a cylinder. Depending on the quenched disorder distribution of the fuse thresholds, we show analytically that the system can either stay in a percolation regime up to breakdown, or start at some current level to localize starting from the smallest scale (lattice spacing), or instead go to a diffuse localization regime where damage starts to concentrate in bands of width scaling as the width of the system, but remains diffuse at smaller scales. Depending on the nature of the quenched disorder on the fuse thresholds, we derive analytically the phase diagram of the system separating these regimes and the current levels for the onset of these possible localizations. We compare these predictions to numerical results.
Although high-transmission-capacity optical fibers are in demand, the problem of the fiber fuse phenomenon needs to be resolved to prevent the destruction of fibers. As polymer optical fibers become more prevalent, clarifying their fuse properties has become important. Here, we experimentally demonstrate a fuse propagation velocity of 21.9 mm/s, which is 1 to 2 orders of magnitude slower than that in standard silica fibers. The achieved threshold power density and proportionality constant between the propagation velocity and the power density are respectively 1/186 of and 16.8 times the values for silica fibers. An oscillatory continuous curve instead of periodic voids is formed after the passage of the fuse. An easy fuse termination method is presented herein, along with its potential plasma applications.
We have used observations taken under the FUSE S405/S505 channel realignment program to explore the diffuse FUV (1000 - 1200 Å) radiation field. Of the 71 independent locations in that program, we have observed a diffuse signal in 32, ranging in brightness from 1600 photons cm$^{-2}$ s$^{-1}$ sr$^{-1}$ Å$^{-1}$ to a maximum of $2.9 \times 10^5$ photons cm$^{-2}$ s$^{-1}$ sr$^{-1}$ Å$^{-1} in Orion. The FUSE data confirm that the diffuse FUV sky is patchy with regions of intense emission, usually near bright stars, but also with dark regions, even at low Galactic latitudes. We find a weak correlation between the FUV flux and the 100 micron ratio but with wide variations, perhaps due to differences in the local radiation field.
We present heavy element abundance measurements for 16 DAO white dwarfs, determined from Far-Ultraviolet Spectroscopic Explorer {FUSE} spectra. Evidence of absorption by heavy elements was found in the spectra of all the objects. Measurements were made using models that adopted the temperatures, gravities and helium abundances determined from both optical and FUSE data by Good et al. (2004). Despite the different evolutionary paths that the types of DAO white dwarfs are thought to evolve through, their abundances were not found to vary significantly, apart from for the silicon abundances. Abundances measured when the FUSE derived values of temperature, gravity and helium abundance were adopted were, in general, a factor 1-10 higher than those determined when the optical measure of those parameters was used. Satisfactory fits to the absorption lines were achieved in approximately equal number. The models that used the FUSE determined parameters seemed better at reproducing the strength of the nitrogen and iron lines, while for oxygen, the optical parameters were better.
FUSE spectra of the four brightest H II regions in M 33 show absorption by interstellar gas in the Galaxy and in M 33. On three lines of sight molecular hydrogen in M 33 is detected. This is the first measurement of diffuse H_2 in absorption in a Local Group galaxy other than the Magellanic Clouds. A quantitative analysis is difficult because of the low signal to noise ratio and the systematic effects produced by having multiple objects in the FUSE aperture. We use the M 33 FUSE data to demonstrate in a more general manner the complexity of interpreting interstellar absorption line spectra towards multi-object background sources. We derive H_2 column densities of approximately 10^16 to 10^17 cm^{-2} along 3 sight lines (NGC 588, NGC 592, NGC 595). Because of the systematic effects, these values most likely represent upper limits and the non-detection of H_2 towards NGC 604 does not exclude the existence of significant amounts of molecular gas along this sight line.
FUSE observations have been instrumental in advancing our understanding of the physical properties and behavior of active binary systems, including cataclysmic variables (CVs) and X-ray binaries (XRBs). FUSE data have allowed observers to study: accretion disks, magnetically-channeled accretion flow, and white dwarf accretors, and how these respond to accretion fluctuations and disk outbursts; the role of binary evolution in determining system properties; the vertical and azimuthal structure in disks and disk winds; and what the variations in active binary properties reveal about accretion physics in compact systems. Results of FUSE observations of active binaries are reviewed here.
In this work, we present an experimental investigation of the fuse model. Our main goal was to study the influence of the disorder on the fracture process. The experimental apparatus used consisted of an $L\times L$ square lattice with fuses placed on each bond of the lattice. Two types of materials were used as fuses: copper and steel wool wires. The lattice composed only by copper wires varied from a weakly disordered system to a strongly disordered one. The lattice formed only by steel wool wires corresponded to a strongly disordered one. The experimental procedure consisted of applying a potential difference V to the lattice and measuring the respective current I. The characteristic function $I(V)$ obtained was investigated in order to find the scaling law dependence of the voltage and the current on the system size $L$ when the disorder was changed. Our results show that the scaling laws are only verified for the disordered regime.
We describe a moderate-resolution FUSE mini-survey of H2 in the Milky Way and Magellanic Clouds, using four hot stars and four AGN as background sources. FUSE spectra of nearly every stellar and extragalactic source exhibit numerous absorption lines from the H2 Lyman and Werner bands between 912 and 1120 A. One extragalactic sightline, PKS 2155-304, with low N(HI) shows no detectable H2 and could be the "Lockman Hole" of molecular gas, of importance for QSO absorption-line studies. We measure H2 column densities in low rotational states (J = 0 and 1) to derive rotational and/or kinetic temperatures of diffuse interstellar gas. The higher-J abundances can constrain models of the UV radiation fields and gas densities. In three optically thick clouds toward extragalactic sources, we find n(H) ~ 30-50 cm(-3) and cloud thicknesses of 2-3 pc. The rotational temperatures for H2 at high Galactic latitude, <T_01> = 107 +/- 17 K (seven sightlines) and 120 +/- 13 K (three optically thick clouds), are higher than those in the Copernicus sample composed primarily of targets in the disk. We find no evidence for great differences in the abundance or state of excitation of H2 between sight l
We present far-ultraviolet observations of IC 63, an emission/reflection nebula illuminated by the B0.5IV star gamma Cassiopeia, located 1.3 pc from the nebula. Molecular hydrogen fluorescence was detected first in IC 63 by IUE and later at shorter wavelengths by ORFEUS. Here we present Far Ultraviolet Spectroscopic Explorer (FUSE) observations towards three locations in the nebula, complemented by Hopkins Ultraviolet Telescope (HUT) data on the central nebular position. In addition, we present a sounding rocket calibration of a FUSE spectrum of gamma Cas. Molecular hydrogen fluorescence is detected in all three FUSE pointings. The intensity of this emission as well as the contributions from other species are seen to vary with position. The absolute flux calibration of the sounding rocket data allows us to reliably predict the radiation field incident on IC 63. We use these data to test models of the fluorescent process. Our modeling resolves the perceived discrepancy between the existing ultraviolet observations and achieves a satisfactory agreement with the H_2 rotational structure observed with FUSE.
The concept of flow is used extensively in HCI, video games, and many other fields, but its prevalent definition is conceptually vague and alternative interpretations have contributed to ambiguity in the literature. To address this, we use cognitive science theory to expose inconsistencies in flow's prevalent definition, and introduce fuse, a concept related to flow but consistent with cognitive science, and defined as the "fusion of activity-related sensory stimuli and awareness". Based on this definition, we develop a preliminary model that hypothesizes fuse's underlying cognitive processes. To illustrate the model's practical value, we derive a set of design heuristics that we exemplify in the context of video games. Together, the fuse definition, model and design heuristics form our theoretical framework, and are a product of rethinking flow from a cognitive perspective with the purpose of improving conceptual clarity and theoretical robustness in the literature.