Scientific document representation learning provides powerful embeddings for various tasks, while current methods face challenges across three approaches. 1) Contrastive training with citation-structural signals underutilizes citation information and still generates single-vector representations. 2) Fine-grained representation learning, which generates multiple vectors at the sentence or aspect level, requires costly integration and lacks domain generalization. 3) Task-aware learning depends on manually predefined task categorization, overlooking nuanced task distinctions and requiring extra training data for task-specific modules. To address these problems, we propose a new method that unifies the three approaches for better representations, namely FLeW. Specifically, we introduce a novel triplet sampling method that leverages citation intent and frequency to enhance citation-structural signals for training. Citation intents (background, method, result), aligned with the general structure of scientific writing, facilitate a domain-generalized facet partition for fine-grained representation learning. Then, we adopt a simple weight search to adaptively integrate three facet-level em
We expand FLew with a unary connective whose algebraic counterpart is the operation that gives the greatest complemented element below a given argument. We prove that the expanded logic is conservative and has the Finite Model Property. We also prove that the corresponding expansion of the class of residuated lattices is an equational class.
Here we present an updated shape model of (486958) Arrokoth, the bilobate Kuiper Belt Object (KBO) which the NASA New Horizons spacecraft flew past in 2019. This updated shape model uses all of the resolved images of Arrokoth obtained by the New Horizons LOng Range Reconnaissance Imager (LORRI). We developed an updated shape modeling algorithm which allowed the shape and rotational pole of Arrokoth to be fit to much better quality with an efficient use of GPU-accelerated features. The resulting model of Arrokoth's contact binary shape is significantly thicker and of larger volume than the one previously published immediately after the flyby by Spencer et al (2020). We show that Arrokoth's smaller lobe Weeyo is roughly spherical in shape, while the larger lobe Wenu is more flattened, with the volume ratio between the lobes being roughly 2:1. Owing to Wenu's oblate shape, Arrokoth's rotational lightcurve would have significantly lower mean reflectance when viewed from subobserver latitudes that would have shown lightcurve variation. We discuss the impact this may have on estimates of the frequency of contact binaries in the Kuiper Belt. We also discuss the implications of this shape
Contrails account for a large portion of aviation's contribution to anthropogenic climate change. Navigational contrail avoidance is a promising solution to mitigate the warming caused by contrails. Prior trials testing navigational contrail avoidance have relied on bespoke integrations of contrail forecasts into airline operations. Here, we use a randomized control trial to test the feasibility of dispatcher-led contrail avoidance integrated into standard flight planning operations using a workflow that scales to an airline's entire network. We validated the efficacy of this intervention using satellite imagery and an automated flight-contrail attribution algorithm. Using this system, we observed an 11.6% reduction in contrail formation rate for the 1232 flights marked as eligible for contrail avoidance (intent-to-treat) relative to the flights in the control group (p = 0.011). In the 112 flights that flew contrail avoidance as planned (per-protocol flights), we observed a 62.0% lower contrail formation rate relative to the flights in the control group (p < 0.001). No statistically significant difference in fuel usage was observed between the two groups.
The Extreme Universe Space Observatory on a Super Pressure Balloon 2 (EUSO-SPB2) flew in May of 2023, marking an important step towards the observation of ultra-high-energy cosmic rays (UHECR) and neutrino-induced showers from space. The ultimate goal of this endeavor is to complement ground-based detectors and achieve unprecedented exposure and nearly uniform full-sky coverage at the highest energies, thereby enabling charged particle astronomy and enriching the multi-messenger approach to high-energy astrophysics and astroparticle physics. As a pathfinder to the POEMMA mission (Probe Of Extreme Multi-Messenger Astrophysics), EUSO-SPB2 flew two distinct cameras at the focus of two Schmidt telescopes, one made of multi-anode photomultiplier tubes (MAPMTs), looking towards the nadir for fluorescence light detection, the other made of Silicon photomultipliers (SiPMs), looking towards the limb of the Earth for direct Cherenkov light detection. The flight was terminated prematurely due to a failure in the balloon, and thus no showers were detected in the fluorescence mode. However, several lower-energy (PeV scale) cosmic-ray events were observed in the Cherenkov channel. The data colle
Quantifying the collision risk of birds and bats with offshore wind turbines requires an understanding of the drivers of flying animal behavior at offshore wind sites. An omnidirectional S-band radar system was deployed on a research barge on the Northeastern Shelf of the United States (40.9 deg N, 70.79 deg W) and collected data for a 5-week window during the 2024 autumn bird and bat migration. The barge also supported two profiling lidar systems that measured the wind speed and direction. This study presents a first methodological approach for analyzing radar and lidar data together, providing a framework for future analyses of offshore bird and bat movements that can be used to improve collision risk models. Coupling the radar animal tracks with measured wind speed profiles revealed that wind is a driver of animal presence, flight direction, flight height, and flight speed. Further, a hierarchical clustering methodology was developed to investigate behavior by approximate animal size. For example, smaller animals had concentrated flight direction distributions aligned with the wind and flew at a variety of altitudes, whereas bigger animals flew in a wide variety of directions bu
On November 6, 2024, the Parker Solar Probe flew past Venus to make the first accurate electric field measurement in the nightside Venusian magnetosphere. To achieve this result, the electric field antennas were current biased in a way never before experienced by an electric field detector. This biasing requirement, that the positive bias current in the Venus shadow be about equal to the electron thermal current, is discussed and illustrated. About one minute of useful electric f ield data in the eight-minute nightside magnetosphere crossing was obtained, during which the only feature observed was a few Hz signal. This result, along with the magnetic field measurements, showed that there were few if any electromagnetic waves, such as low-frequency electromagnetic turbulence or whistlers, in the nightside crossing. Instead, a few Hertz, purely electrostatic signal was found. This suggests that the interaction of the solar wind with an unmagnetized body having an ionosphere may be different from that of previously studied magnetized bodies. In the sunlit flanks, many electromagnetic wave modes were observed. These results describe the first step in the proper technique for future mea
3I/ATLAS is the third confirmed interstellar object passing through the Solar System. In this work, we conduct narrowband radio technosignature search toward 3I/ATLAS using the Five-hundred-meter Aperture Spherical Telescope (FAST) L-band multibeam receiver from October 2025 to January 2026 on 4 separate dates (i.e. Mars closest, perihelion, Earth closest and flew away from Earth, respectively). We carry out frequency-drifting signal searching with signal-to-noise ratio (SNR) over 10 within 1.05-1.45 GHz via \texttt{bliss} pipeline. These signal hits are grouping into event by beam, frequency and drift rate matching, the events are then filtered by cluster analysis and drift rate cut-off. We also characterized the events by their significance in SNR, structure tensor as well as principal component analysis (PCA). No credible narrowband radio technosignature are detected from 3I/ATLAS after visual inspections. The null results place constraints on the presence of transmitters above $2.862\times 10^{-3}$ W. We further introduce a Bayesian inference framework to constraints on the existence probability and characteristic power of hypothetical transmitters using physically motivated pr
This study compares aerodynamic performance of the CLF5605 rotor airfoil -- which flew on Ingenuity from 2021 to 2024 -- with that of a new optimized roamx-0201 airfoil designed for Martian conditions at NASA Ames. Specifically, performance is studied at a Reynolds number of 20,000 and a Mach number of 0.60, across a range of angles of attack, using three independent state-of-the-art methodologies: implicit large eddy simulations (ILES) using NASA's OVERFLOW solver, direct numerical simulations (DNS) using the high-order GPU-accelerated PyFR solver, and experimental testing in the Mars Wind Tunnel at Tohoku University. Discrepancies between results obtain using the various methodologies are analyzed and explained. Across all methodologies it can be seen that the roamx-0201 airfoil is able to achieve a given lift with less drag compared to the CLF5605 airfoil. Moreover, OVERFLOW and PyFR results show that the roamx-0201 airfoil has superior stall characteristics, and can achieve a maximum lift ~20% higher than that achieved by the CLF5605 airfoil. The work provides a strong body of evidence to support further studies into use of rotors based on the optimized roamx-0201 airfoil for f
We present an updated atlas of spectra from the Low-Resolution Spectrometer (LRS) on the Infrared Astronomical Satellite (IRAS), which took spectra from 7.67 to 22.73 um with a spectral resolving power (lambda / Delta lambda) of 20-60. The updated atlas includes 11,238 spectra, including 5425 spectra published in the original LRS Atlas, 5796 spectra published in three later papers, and 17 spectra previously available online but not published. The updated atlas has significantly more sources close to the Galactic plane than the original atlas. We have applied an improved spectral correction to remove an artifact at 8 um in the original database. While the IRAS mission flew over 40 yr ago, the extended LRS atlas remains the single most complete database of mid-infrared spectra of nearby and bright objects in the Galaxy.
Spectroscopy is the most important method for probing the structure of molecules. However, predicting molecular spectra on classical computers is computationally expensive, with the most accurate methods having a cost that grows exponentially with molecule size. Quantum computers have been shown to simulate simple types of optical spectroscopy efficiently -- with a cost polynomial in molecule size -- using methods such as time-dependent simulations of photoinduced wavepackets. Here, we show that any type of spectroscopy can be efficiently simulated on a quantum computer using a time-domain approach, including spectroscopies of any order, any frequency range, and involving both electric and magnetic transitions. Our method works by computing any spectroscopic correlation function based on the corresponding double-sided Feynman diagram, the canonical description of spectroscopic interactions. The approach can be used to simulate spectroscopy of both closed and open molecular systems using both digital and analog quantum computers.
The Focusing Optics X-ray Solar Imager (FOXSI) sounding rocket payload flew for the second time on 2014 December 11. To enable direct Hard X-Ray (HXR) imaging spectroscopy, FOXSI makes use of grazing-incidence replicated focusing optics combined with fine-pitch solid-state detectors. FOXSI's first flight provided the first HXR focused images of the Sun. For FOXSI's second flight several updates were made to the instrument including updating the optics and detectors as well as adding a new Solar Aspect and Alignment System (SAAS). This paper provides an overview of these updates as well as a discussion of their measured performance.
Cosmic Microwave Background experiments need to measure polarization properties of the incoming radiation very accurately to achieve their scientific goals. As a result of that, it is necessary to properly characterize these instruments. However, there are not natural sources that can be used for this purpose. For this reason, we developed the PROTOtype CALibrator for Cosmology, PROTOCALC, which is a calibrator source designed for the 90 GHz band of these telescopes. This source is purely polarized and the direction of the polarization vector is known with an accuracy better than 0.1 deg. This source flew for the first time in May 2022 showing promising result.
This paper details a significant milestone towards maturing a buoyant aerial robotic platform, or aerobot, for flight in the Venus clouds. We describe two flights of our subscale altitude-controlled aerobot, fabricated from the materials necessary to survive Venus conditions. During these flights over the Nevada Black Rock desert, the prototype flew at the identical atmospheric densities as 54 to 55 km cloud layer altitudes on Venus. We further describe a first-principle aerobot dynamics model which we validate against the Nevada flight data and subsequently employ to predict the performance of future aerobots on Venus. The aerobot discussed in this paper is under JPL and Aerostar development for an in-situ mission flying multiple circumnavigations of Venus, sampling the chemical and physical properties of the planet's atmosphere and also remotely sensing surface properties.
Observing ultra-high energy cosmic rays (UHECR) and very high energy (VHE) neutrinos from space is a promising way to measure their extremely low fluxes by significantly increasing the observed volume. The Extreme Universe Space Observatory on a Super Pressure Balloon 2 (EUSO-SPB2), the next, most advanced pathfinder for such a mission, was launched May 13th 2023 from Wanaka New Zealand. The pioneering EUSO-SPB2 payload flew a Fluorescence Telescope (FT) with a PMT camera pointed in nadir to record fluorescence light from cosmic ray extensive air shower (EAS) with energies above 1 EeV, and a Cherenkov telescope (CT) with a silicon photomultiplier focal surface for observing Cherenkov emission of cosmic ray EAS with energies above 1 PeV with an above-the-limb geometry and of PeV-scale EAS initiated by neutrino-sourced tau decay. As the CT is a novel instrument, optical background measurements for space neutrino observation are an important goal of the mission. Any data collected during the mission will influence and improve the development of a space-based multi-messenger observatory such as the Probe of Extreme Multi-Messenger Astrophysics (POEMMA). We present an overview of the EU
The Extreme Universe Space Observatory on a Super Pressure Balloon 2 (EUSO-SPB2), successfully launched from Wanaka, New Zealand on May 13, 2022, is a precursor for a space-based astroparticle observatory such as the Probe Of Extreme Multi-Messenger Astrophysics (POEMMA). EUSO-SPB2 flew two custom telescopes. Both have UV/UV-visible sensitivity and feature Schmidt optics. The Fluorescence Telescope (FT) measures ultra-high energy cosmic rays by looking down. The Čerenkov Telescope (CT) searches for neutrino signatures by looking toward Earth's limb. The two telescopes each have a 1 m diameter entrance pupil and segmented glass mirrors that collect light from extensive air showers at the PeV and EeV-scale. Here we describe the FT telescope optics together with the results of the FT field tests at the Utah Telescope Array (TA) site from August/September 2022. The FT recorded the night sky background, lasers, and artificial point sources. The field tests included an absolute photometric calibration of the FT telescope that is compared to a piece-wise laboratory calibration.
We present MONET, a new multimodal dataset captured using a thermal camera mounted on a drone that flew over rural areas, and recorded human and vehicle activities. We captured MONET to study the problem of object localisation and behaviour understanding of targets undergoing large-scale variations and being recorded from different and moving viewpoints. Target activities occur in two different land sites, each with unique scene structures and cluttered backgrounds. MONET consists of approximately 53K images featuring 162K manually annotated bounding boxes. Each image is timestamp-aligned with drone metadata that includes information about attitudes, speed, altitude, and GPS coordinates. MONET is different from previous thermal drone datasets because it features multimodal data, including rural scenes captured with thermal cameras containing both person and vehicle targets, along with trajectory information and metadata. We assessed the difficulty of the dataset in terms of transfer learning between the two sites and evaluated nine object detection algorithms to identify the open challenges associated with this type of data. Project page: https://github.com/fabiopoiesi/monet_datase
Cosmic Microwave Background experiments need to measure polarization properties of the incoming radiation very accurately to achieve their scientific goals. As a result of that, it is necessary to properly characterize these instruments. However, there are not natural sources that can be used for this purpose. For this reason, we developed the PROTOtype CALibrator for Cosmology, PROTOCALC, which is a calibrator source designed for the 90GHz band of these telescopes. This source is purely polarized and the direction of the polarization vector is known with an accuracy better than 0.1deg. This source flew for the first time in May 2022 showing promising result
The Ulysses spacecraft is orbiting the Sun on a highly inclined ellipse ($i = 79^{\circ}$). After its Jupiter flyby in 1992 at a heliocentric distance of 5.4 AU, the spacecraft reapproached the inner solar system, flew over the Sun's south polar region in September 1994, crossed the ecliptic plane at a distance of 1.3 AU in March 1995, and flew over the Sun's north polar region in July 1995. We report on dust impact data obtained with the dust detector onboard Ulysses between January 1993 and December 1995. We publish and analyse the complete data set of 509 recorded impacts of dust particles with masses between $10^{-16}$ g to $10^{-7}$ g. Together with 968 dust impacts from launch until the end of 1992 published earlier (Grün et al., 1995, {\em Planet. Space Sci}, Vol. 43, p. 971-999), information about 1477 particles detected with the Ulysses sensor between October 1990 and December 1995 is now available. The impact rate measured between 1993 and 1995 stayed relatively constant at about 0.4 impacts per day and varied by less than a factor of ten. Most of the impacts recorded outside about 3.5 AU are compatible with particles of interstellar origin. Two populations of interplanet
Figurative language (e.g., "he flew like the wind") is challenging to understand, as it is hard to tell what implicit information is being conveyed from the surface form alone. We hypothesize that to perform this task well, the reader needs to mentally elaborate the scene being described to identify a sensible meaning of the language. We present DREAM-FLUTE, a figurative language understanding system that does this, first forming a "mental model" of situations described in a premise and hypothesis before making an entailment/contradiction decision and generating an explanation. DREAM-FLUTE uses an existing scene elaboration model, DREAM, for constructing its "mental model." In the FigLang2022 Shared Task evaluation, DREAM-FLUTE achieved (joint) first place (Acc@60=63.3%), and can perform even better with ensemble techniques, demonstrating the effectiveness of this approach. More generally, this work suggests that adding a reflective component to pretrained language models can improve their performance beyond standard fine-tuning (3.3% improvement in Acc@60).