Search

Thanks to the complementary nature of millimeter wave radar and camera, deep learning-based radar-camera 3D object detection methods may reliably produce accurate detections even in low-visibility conditions. This makes them preferable to use in autonomous vehicles' perception systems, especially as the combined cost of both sensors is cheaper than the cost of a lidar. Recent radar-camera methods commonly perform feature-level fusion which often involves projecting the radar points onto the same plane as the image features and fusing the extracted features from both modalities. While performing fusion on the image plane is generally simpler and faster, projecting radar points onto the image plane flattens the depth dimension of the point cloud which might lead to information loss and makes extracting the spatial features of the point cloud harder. We proposed ClusterFusion, an architecture that leverages the local spatial features of the radar point cloud by clustering the point cloud and performing feature extraction directly on the point cloud clusters before projecting the features onto the image plane. ClusterFusion achieved the state-of-the-art performance among all radar-monocular camera methods on the test slice of the nuScenes dataset with 48.7% nuScenes detection score (NDS). We also investigated the performance of different radar feature extraction strategies on point cloud clusters: a handcrafted strategy, a learning-based strategy, and a combination of both, and found that the handcrafted strategy yielded the best performance. The main goal of this work is to explore the use of radar's local spatial and point-wise features by extracting them directly from radar point cloud clusters for a radar-monocular camera 3D object detection method that performs cross-modal feature fusion on the image plane., Comment: Accepted for publication in IEEE Access

Multi-messenger searches for BNS and NSBH mergers are currently one of the most exciting areas of astronomy. The search for joint electromagnetic and neutrino counterparts to GWs has resumed with O4. To support this effort, public semi-automated data products are sent in near real-time and include localization and source properties to guide complementary observations. In preparation for O4, we have conducted a study using a simulated population of compact binaries and a MDC in the form of a real-time replay to optimize and profile the software infrastructure and scientific deliverables. End-to-end performance was tested, including data ingestion, running online search pipelines, performing annotations, and issuing alerts to the astrophysics community. We present an overview of the low-latency infrastructure and the performance of the data products that are now being released during O4 based on the MDC. We report the expected median latency for the preliminary alert of full bandwidth searches (29.5s) and show consistency and accuracy of released data products using the MDC. For the first time, we report the expected median latency for triggers from early warning searches (-3.1s), which are new in O4 and target neutron star mergers during inspiral phase. This paper provides a performance overview for LVK low-latency alert infrastructure and data products using the MDC and serves as a useful reference for the interpretation of O4 detections.

We investigate the dynamics responsible for generating the potential of the $\eta'$, the (would-be) Goldstone boson associated with the anomalous axial $U(1)$ symmetry of QCD. The standard lore posits that pure QCD dynamics generates a confining potential with a branched structure as a function of the $\theta$ angle, and that this same potential largely determines the properties of the $\eta'$ once fermions are included. Here we test this picture by examining a supersymmetric extension of QCD with a small amount of supersymmetry breaking generated via anomaly mediation. For pure $SU(N)$ QCD without flavors, we verify that there are $N$ branches generated by gaugino condensation. Once quarks are introduced, the flavor effects qualitatively change the strong dynamics of the pure theory. For $F$ flavors we find $|N-F|$ branches, whose dynamical origin is gaugino condensation in the unbroken subgroup for $FN+1$. For the special cases of $F = N-1, N, N + 1$ we find no branches and the entire potential is consistent with being a one-instanton effect. The number of branches is a simple consequence of the selection rules of an anomalous $U(1)_R$ symmetry. We find that the $\eta'$ mass does not vanish in the large $N$ limit for fixed $F/N$, since the anomaly is non-vanishing. The same dynamics that is responsible for the $\eta'$ potential is also responsible for the axion potential. We present a simple derivation of the axion mass formula for an arbitrary number of flavors., Comment: 34 pages, 2 figures

Bismuth is a candidate material for three dimensional higher dimensional topological insulators. We performed electronic transport experiments on small diameter crystalline bismuth nanowires to clarify the role of the proposed hinge channels in the interlayer coupling. The magnetoresistance presents a sequence of peaks at Yamaji magic angles for which the interlayer group velocity is zero pointing to flat bands due to the layered structure. Furthermore, we observe a peak for high magnetic fields applied parallel to the layers, a definitive signature of interlayer coherence that enables deduction of the interlayer transfer integral of 8 meV. We demonstrate transport by a corrugated open Fermi surface of holes that is extended, with the high angular precision of around 0.016 rad in the interlayer direction. The observations indicate that coherence and the crystallographic direction set by hinges play a key role in the interlayer coupling of bismuth., Comment: 22 pages, 4 figures

This article deals with an autonomous differential equation model that studies the interaction between the immune system and the growth of tumor cells with strong and weak Allee effects. The Allee effect refers to interspecific competition, and when the population is small, it can retard population growth. The work focuses on describing analytically, using a set of parameters, the conditions in the phases of the immunoediting theory, particularly in the equilibrium phase, where a latent tumor would exist. Saddle-Node, Saddle-symmetric, Hopf, generalized Hopf, and Takens-Bogdanov bifurcations get presented for both Allee effects, and their biological interpretation regarding cancer dynamics gets discussed. The Hopf and generalized Hopf bifurcation curves get analyzed through hyper-parameter projections of the model, where it gets observed that with a strong Allee effect, more tumor control persists as it has higher antigenicity, in contrast to the weak Allee effect, where lower antigenicity gets observed. Also, we observe that the equilibrium phase persists as antigenicity increases with a strong Allee effect. Finally, the numerical continuation gets performed to replicate the analytical curves' bifurcations and draw the limit and double limit cycles.

Higgs coupling deviations from Standard Model predictions contain information about two scales of Nature: that of new physics responsible for the deviation, and the scale where new bosons must appear. The two can coincide, but they do not have to. The scale of new bosons can be calculated by going beyond an effective field theory description of the coupling deviation. We compute model-independent upper bounds on the scale of new bosons for deviations in Higgs to $WW$ and $ZZ$ couplings, finding that any measured deviation at present or future colliders requires the existence of new bosons within experimental reach. This has potentially interesting implications for naturalness., Comment: 41 pages + references, 14 figures. v2: JHEP version

We call on the Document AI (DocAI) community to reevaluate current methodologies and embrace the challenge of creating more practically-oriented benchmarks. Document Understanding Dataset and Evaluation (DUDE) seeks to remediate the halted research progress in understanding visually-rich documents (VRDs). We present a new dataset with novelties related to types of questions, answers, and document layouts based on multi-industry, multi-domain, and multi-page VRDs of various origins, and dates. Moreover, we are pushing the boundaries of current methods by creating multi-task and multi-domain evaluation setups that more accurately simulate real-world situations where powerful generalization and adaptation under low-resource settings are desired. DUDE aims to set a new standard as a more practical, long-standing benchmark for the community, and we hope that it will lead to future extensions and contributions that address real-world challenges. Finally, our work illustrates the importance of finding more efficient ways to model language, images, and layout in DocAI., Comment: Accepted at ICCV 2023

We discuss a novel detection technique for millicharged dark matter that makes use of existing light-shining-through-wall (LSW) experiments searching for massive dark photons. Since millicharged particles interact with both the visible and dark sectors, a small background of such particles enables the search for visible signals even in the limit of a massless dark photon. Furthermore, for sufficiently large couplings, a small dark matter subcomponent consisting of millicharged particles can have a terrestrial abundance that is significantly enhanced with respect to its galactic abundance. We leverage this fact to show that in certain parts of parameter space, future runs of the LSW Dark SRF experiment can be used to set the strongest limits on millicharged relics., Comment: 17 pages, 4 figures