Search

Open-vocabulary 3D instance segmentation transcends traditional closed-vocabulary methods by enabling the identification of both previously seen and unseen objects in real-world scenarios. It leverages a dual-modality approach, utilizing both 3D point clouds and 2D multi-view images to generate class-agnostic object mask proposals. Previous efforts predominantly focused on enhancing 3D mask proposal models; consequently, the information that could come from 2D association to 3D was not fully exploited. This bias towards 3D data, while effective for familiar indoor objects, limits the system's adaptability to new and varied object types, where 2D models offer greater utility. Addressing this gap, we introduce Zero-Shot Dual-Path Integration Framework that equally values the contributions of both 3D and 2D modalities. Our framework comprises three components: 3D pathway, 2D pathway, and Dual-Path Integration. 3D pathway generates spatially accurate class-agnostic mask proposals of common indoor objects from 3D point cloud data using a pre-trained 3D model, while 2D pathway utilizes pre-trained open-vocabulary instance segmentation model to identify a diverse array of object proposals from multi-view RGB-D images. In Dual-Path Integration, our Conditional Integration process, which operates in two stages, filters and merges the proposals from both pathways adaptively. This process harmonizes output proposals to enhance segmentation capabilities. Our framework, utilizing pre-trained models in a zero-shot manner, is model-agnostic and demonstrates superior performance on both seen and unseen data, as evidenced by comprehensive evaluations on the ScanNet200 and qualitative results on ARKitScenes datasets., Comment: OpenSUN 3D: 2nd Workshop on Open-Vocabulary 3D Scene Understanding (CVPR 2024)

Spatio-temporal distribution of urban population is crucial to understand the structure and dynamics of cities. Most studies, however, have focused on the microscopic structure of cities such as their few most crowded areas. In this work, we investigate the macroscopic structure of cities such as their clusters of highly populated areas. To do this, we analyze the spatial distribution of urban population and its intra-day dynamics in Seoul and Helsinki with a percolation framework. We observe that the growth patterns of the largest clusters in the real and randomly shuffled population data are significantly different, and highly populated areas during daytime are denser and form larger clusters than during nighttime. An analysis of the cluster size distributions at percolation criticality shows that their power-law exponents during daytime are lower than during nighttime, indicating that the spatial distributions of urban population during daytime and nighttime fall into different universality classes. Finally measuring the area-perimeter fractal dimension of the collection of clusters at criticality demonstrates that the fractal dimensions during daytime are higher than during nighttime, indicating that the perimeters of clusters during daytime are more rough than during nighttime. Our findings suggest that even the same city can have qualitatively different spatial distributions of population over time, and propose a way to quantitatively compare the macrostructure of cities based on population distribution data., Comment: 9 pages, 6 figures. Submitted for publication

Unconventional spin-orbit torques arising from electric-field-generated spin currents in anisotropic materials have promising potential for spintronic applications, including for perpendicular magnetic switching in high-density memory applications. Here we determine all the independent elements of the spin torque conductivity tensor allowed by bulk crystal symmetries for the tetragonal conductor IrO2, via measurements of conventional (in plane) antidamping torques for IrO2 thin films in the high-symmetry (001) and (100) orientations. We then test that rotational transformations of this same tensor can predict both the conventional and unconventional anti-damping torques for IrO2 thin films in the lower-symmetry (101), (110), and (111) orientations, finding good agreement. The results confirm that spin-orbit torques from all these orientations are consistent with the bulk symmetries of IrO2, and show how simple measurements of conventional torques from high-symmetry orientations of anisotropic thin films can provide an accurate prediction of the unconventional torques from lower-symmetry orientations., Comment: Main text: 14 pages, 3 figures. Supporting Information: 19 pages, 10 figures, and 1 table

Identifying influential spreaders is a crucial problem for practical applications in network science. The core-periphery(C-P) structure, common in many real-world networks, comprises a densely interconnected group of nodes(core) and the rest of the sparsely connected nodes subordinated to the core(periphery). Core nodes are expected to be more influential than periphery nodes generally, but recent studies suggest that this is not the case in some networks. In this work, we look for mesostructural conditions that arise when core nodes are significantly more influential than periphery nodes. In particular, we investigate the roles of the internal and external connectivity of cores in their relative influence. We observe that the internal and external connectivity of cores are broadly distributed, and the relative influence of the cores is also broadly distributed in real-world networks. Our key finding is that the internal connectivity of cores is positively correlated with their relative influence, whereas the relative influence increases up to a certain value of the external connectivity and decreases thereafter. Finally, results from the model-generated networks clarify the observations from the real-world networks. Our findings provide a structural condition for influential cores in networks and shed light on why some cores are influential and others are not.

Reinforcement Learning from Human Feedback (RLHF) leverages human preference data to train language models to align more closely with human essence. These human preference data, however, are labeled at the sequence level, creating a mismatch between sequence-level preference labels and tokens, which are autoregressively generated from the language model. Although several recent approaches have tried to provide token-level (i.e., dense) rewards for each individual token, these typically rely on predefined discrete reward values (e.g., positive: +1, negative: -1, neutral: 0), failing to account for varying degrees of preference inherent to each token. To address this limitation, we introduce TLCR (Token-Level Continuous Reward) for RLHF, which incorporates a discriminator trained to distinguish positive and negative tokens, and the confidence of the discriminator is used to assign continuous rewards to each token considering the context. Extensive experiments show that our proposed TLCR leads to consistent performance improvements over previous sequence-level or token-level discrete rewards on open-ended generation benchmarks., Comment: ACL2024 Findings

We present the design and experimental characterization of a kinetic-inductance traveling-wave parametric amplifier (KI-TWPA) for sub-GHz frequencies. KI-TWPAs amplify signals through nonlinear mixing processes supported by the nonlinear kinetic inductance of a superconducting transmission line. The device described here utilizes a compactly meandered TiN microstrip transmission line to achieve the length needed to amplify sub-GHz signals. It is operated in a frequency translating mode where the amplified signal tone is terminated at the output of the amplifier, and the idler tone at approximately 2.5~GHz is brought out of the cryostat. By varying the pump frequency, a gain of up to 22 dB was achieved in a tunable range from about 450 to 850~MHz. Use of TiN as the nonlinear element allows for a reduction of the required pump power by roughly an order of magnitude relative to NbTiN, which has been used for previous KI-TWPA implementations. This amplifier has the potential to enable high-sensitivity and high-speed measurements in a wide range of applications, such as quantum computing, astrophysics, and dark matter detection.

On March 13, 2024, Ethereum implemented EIP-4844, designed to enhance its role as a data availability layer. While this upgrade reduces data posting costs for rollups, it also raises concerns about its impact on the consensus layer due to increased propagation sizes. Moreover, the broader effects on the overall Ethereum ecosystem remain largely unexplored. In this paper, we conduct an empirical analysis of the impact of EIP-4844 on consensus security, Ethereum usage, rollup transaction dynamics, and the blob gas fee mechanism. We explore changes in synchronization times, provide quantitative assessments of rollup and user behaviors, and deepen the understanding of the blob gas fee mechanism, highlighting both enhancements and areas of concern post-upgrade.

Bilateral teleoperation of an aerial manipulator facilitates the execution of industrial missions thanks to the combination of the aerial platform's maneuverability and the ability to conduct complex tasks with human supervision. Heretofore, research on such operations has focused on flying without any physical interaction or exerting a pushing force on a contact surface that does not involve abrupt changes in the interaction force. In this paper, we propose a human reaction time compensating haptic-based bilateral teleoperation strategy for an aerial manipulator extracting a wedged object from a static structure (i.e., plug-pulling), which incurs an abrupt decrease in the interaction force and causes additional difficulty for an aerial platform. A haptic device composed of a 4-degree-of-freedom robotic arm and a gripper is made for the teleoperation of aerial wedged object-extracting tasks, and a haptic-based teleoperation method to execute the aerial manipulator by the haptic device is introduced. We detect the extraction of the object by the estimation of the external force exerted on the aerial manipulator and generate reference trajectories for both the aerial manipulator and the haptic device after the extraction. As an example of the extraction of a wedged object, we conduct comparative plug-pulling experiments with a quadrotor-based aerial manipulator. The results validate that the proposed bilateral teleoperation method reduces the overshoot in the aerial manipulator's position and ensures fast recovery to its initial position after extracting the wedged object., Comment: to be presented in 2024 IEEE International Conference on Unmanned Aircraft Systems (ICUAS), Chania, Crete, Greece, 2024

We report measurements characterizing the performance of a kinetic inductance detector array designed for a wavelength of 25 microns and very low optical background level suitable for applications such as a far-infrared instrument on a cryogenically cooled space telescope. In a pulse counting mode of operation at low optical flux, the detectors can resolve individual 25-micron photons. In an integrating mode, the detectors remain photon noise limited over more than six orders of magnitude in absorbed power from 70 zW to 200 fW, with a limiting NEP of 4.6 x 10^-20 W/rtHz at 1 Hz. In addition, the detectors are highly stable with flat power spectra under optical load down to 1 mHz. Operational parameters of the detector are determined including the efficiency of conversion of the incident optical power into quasiparticles in the aluminum absorbing element and the quasiparticle self-recombination constant., Comment: 35 pages, 18 figures

Offline reinforcement learning has emerged as a promising technology by enhancing its practicality through the use of pre-collected large datasets. Despite its practical benefits, most algorithm development research in offline reinforcement learning still relies on game tasks with synthetic datasets. To address such limitations, this paper provides autonomous driving datasets and benchmarks for offline reinforcement learning research. We provide 19 datasets, including real-world human driver's datasets, and seven popular offline reinforcement learning algorithms in three realistic driving scenarios. We also provide a unified decision-making process model that can operate effectively across different scenarios, serving as a reference framework in algorithm design. Our research lays the groundwork for further collaborations in the community to explore practical aspects of existing reinforcement learning methods. Dataset and codes can be found in https://sites.google.com/view/ad4rl., Comment: ICRA 2024 Website at: https://sites.google.com/view/ad4rl

The LaAlO$_3$/SrTiO$_3$ interface hosts a plethora of gate-tunable electronic phases. Gating of LaAlO$_3$/SrTiO$_3$ interfaces are usually assumed to occur electrostatically. However, increasing evidence suggests that non-local interactions can influence and, in some cases, dominate the coupling between applied gate voltages and electronic properties. Here, we sketch quasi-1D ballistic electron waveguides at the LaAlO$_3$/SrTiO$_3$ interface as a probe to understand how gate tunability varies as a function of spatial separation. Gate tunability measurements reveal the scaling law to be at odds with the pure electrostatic coupling observed in traditional semiconductor systems. The non-Coulombic gating at the interface is attributed to the existence of a long-range nanoelectromechanical coupling between the gate and electron waveguide, mediated by the ferroelastic domains in SrTiO$_3$. The long-range interactions at the LaAlO$_3$/SrTiO$_3$ interface add unexpected richness and complexity to this correlated electron system.

In Automatic Speech Recognition (ASR) systems, a recurring obstacle is the generation of narrowly focused output distributions. This phenomenon emerges as a side effect of Connectionist Temporal Classification (CTC), a robust sequence learning tool that utilizes dynamic programming for sequence mapping. While earlier efforts have tried to combine the CTC loss with an entropy maximization regularization term to mitigate this issue, they employed a constant weighting term on the regularization during the training, which we find may not be optimal. In this work, we introduce Adaptive Maximum Entropy Regularization (AdaMER), a technique that can modulate the impact of entropy regularization throughout the training process. This approach not only refines ASR model training but ensures that as training proceeds, predictions display the desired model confidence.

Terahertz (THz) spin waves or their quanta, magnons, can be efficiently excited by acoustic phonons because these excitations have similar wavevectors in the THz regime. THz acoustic phonons can be produced using photoacoustic phenomena but typically have a low population and thus a relatively low displacement amplitude. The magnetization amplitude and population of the acoustically excited THz magnons are thus usually small. Using analytical calculations and dynamical phase-field simulations, we show that a freestanding metal/magnetic-insulator (MI)/dielectric multilayer can be designed to produce large-amplitude THz spin wave via cavity-enhanced magnon-phonon interaction. The amplitude of the acoustically excited THz spin wave in the freestanding multilayer is predicted to be more than ten times larger than in a substrate-supported multilayer. Acoustically excited nonlinear magnon-magnon interaction is demonstrated in the freestanding multilayer. The simulations also indicate that the magnon modes can be detected by probing the charge current in the metal layer generated via spin-charge conversion across the MI/metal interface and the resulting THz radiation. Applications of the freestanding multilayer in THz optoelectronic transduction are computationally demonstrated.

Recently, with increasing interest in pet healthcare, the demand for computer-aided diagnosis (CAD) systems in veterinary medicine has increased. The development of veterinary CAD has stagnated due to a lack of sufficient radiology data. To overcome the challenge, we propose a generative active learning framework based on a variational autoencoder. This approach aims to alleviate the scarcity of reliable data for CAD systems in veterinary medicine. This study utilizes datasets comprising cardiomegaly radiograph data. After removing annotations and standardizing images, we employed a framework for data augmentation, which consists of a data generation phase and a query phase for filtering the generated data. The experimental results revealed that as the data generated through this framework was added to the training data of the generative model, the frechet inception distance consistently decreased from 84.14 to 50.75 on the radiograph. Subsequently, when the generated data were incorporated into the training of the classification model, the false positive of the confusion matrix also improved from 0.16 to 0.66 on the radiograph. The proposed framework has the potential to address the challenges of data scarcity in medical CAD, contributing to its advancement.

Fuzzing is an effective bug-finding technique but it struggles with complex systems like JavaScript engines that demand precise grammatical input. Recently, researchers have adopted language models for context-aware mutation in fuzzing to address this problem. However, existing techniques are limited in utilizing coverage guidance for fuzzing, which is rather performed in a black-box manner. This paper presents a novel technique called CovRL (Coverage-guided Reinforcement Learning) that combines Large Language Models (LLMs) with reinforcement learning from coverage feedback. Our fuzzer, CovRL-Fuzz, integrates coverage feedback directly into the LLM by leveraging the Term Frequency-Inverse Document Frequency (TF-IDF) method to construct a weighted coverage map. This map is key in calculating the fuzzing reward, which is then applied to the LLM-based mutator through reinforcement learning. CovRL-Fuzz, through this approach, enables the generation of test cases that are more likely to discover new coverage areas, thus improving vulnerability detection while minimizing syntax and semantic errors, all without needing extra post-processing. Our evaluation results indicate that CovRL-Fuzz outperforms the state-of-the-art fuzzers in terms of code coverage and bug-finding capabilities: CovRL-Fuzz identified 48 real-world security-related bugs in the latest JavaScript engines, including 39 previously unknown vulnerabilities and 11 CVEs., Comment: 14 pages, 4 figures, 9 tables, 2 listings

Kinetic inductance traveling-wave parametric amplifiers (KI-TWPA) have a wide instantaneous bandwidth with near quantum-limited performance and a relatively high dynamic range. Because of this, they are suitable readout devices for cryogenic detectors and superconducting qubits and have a variety of applications in quantum sensing. This work discusses the design, fabrication, and performance of a KI-TWPA based on four-wave mixing in a NbTiN microstrip transmission line. This device amplifies a signal band from 4 to 8~GHz without contamination from image tones, which are produced in a separate higher frequency band. The 4 - 8~GHz band is commonly used to read out cryogenic detectors, such as microwave kinetic inductance detectors (MKIDs) and Josephson junction-based qubits. We report a measured maximum gain of over 20 dB using four-wave mixing with a 1-dB gain compression point of -58 dBm at 15 dB of gain over that band. The bandwidth and peak gain are tunable by adjusting the pump-tone frequency and power. Using a Y-factor method, we measure an amplifier-added noise of $ 0.5 \leq N_{added} \leq 1.5$ photons from 4.5 - 8 GHz.

Microwave Kinetic Inductance Detectors (MKIDs) have been demonstrated as capable phonon sensors when coupled to crystalline substrates, and have been proposed as detectors for next-generation rare-event searches such as for the direct detection of dark matter. These Kinetic Inductance Phonon Mediated (KIPM) detector designs, favoring large superconducting absorber volumes and high readout powers, are oftentimes limited in their sensitivity by low temperature amplifier noise introduced in the signal readout chain. We report here an effort to couple a wideband Kinetic Inductance Travelling Wave Parametric Amplifier (KI-TWPA), operated near the Standard Quantum Limit of minimal added amplifier noise, to sensors spanning a 70 MHz bandwidth at 3.5 GHz. This results in a ~5x improvement in the inferred detector energy resolution in the best sensor and highlights the potential of constructing O(100) meV resolving phonon-mediated particle detectors. We detail limitations introduced by lossy passive components, degraded RF responsivity, and microphysical noise sources like two-level systems (TLS), in achieving ultimate quantum-limited system noise levels., Comment: 7 pages, 4 figures

KTaO3 has recently attracted attention as a model system to study the interplay of quantum paraelectricity, spin-orbit coupling, and superconductivity. However, the high and low vapor pressures of potassium and tantalum present processing challenges to creating interfaces clean enough to reveal the intrinsic quantum properties. Here, we report superconducting heterostructures based on electronic-grade epitaxial (111) KTaO3 thin films. Electrical and structural characterizations reveal that two-dimensional electron gas at the heterointerface between amorphous LaAlO3 and KTaO3 thin film exhibits significantly higher electron mobility, superconducting transition temperature and critical current density than those in bulk single crystal KTaO3-based heterostructures owing to cleaner interface in KTaO3 thin films. Our hybrid approach may enable epitaxial growth of other alkali metal-based oxides that lie beyond the capabilities of conventional methods.

The Balloon Experiment for Galactic INfrared Science (BEGINS) is a concept for a sub-orbital observatory that will operate from $\lambda$ = 25-250 $\mu$m to characterize dust in the vicinity of high-mass stars. The mission's sensitivity requirements will be met by utilizing arrays of 1,840 lens-coupled, lumped-element kinetic inductance detectors (KIDs) operating at 300 mK. Each KID will consist of a titanium nitride (TiN) parallel strip absorbing inductive section and parallel plate capacitor (PPC) deposited on a silicon (Si) substrate. The PPC geometry allows for reduction of the pixel spacing. At the BEGINS focal plane the detectors require optical NEPs from $2\times10^{-16}$ W/$\sqrt{\textrm{Hz}}$ to $6\times10^{-17}$ W/$\sqrt{\textrm{Hz}}$ from 25-250 $\mu$m for optical loads ranging from 4 pW to 10 pW. We present the design, optical performance and quasiparticle lifetime measurements of a prototype BEGINS KID array at 25 $\mu$m when coupled to Fresnel zone plate lenses. For our optical set up and the absorption efficiency of the KIDs, the electrical NEP requirement at 25 $\mu$m is $7.6\times10^{-17}$ W/$\sqrt{\textrm{Hz}}$ for an absorbed optical power of 0.36 pW. We find that over an average of five resonators the the detectors are photon noise limited down to about 200 fW, with a limiting NEP of about $7.4\times10^{-17}$ W/$\sqrt{\textrm{Hz}}$.

Future space-based far-infrared astrophysical observatories will require exquis-itely sensitive detectors consistent with the low optical backgrounds. The PRobe far-Infrared Mission for Astrophysics (PRIMA) will deploy arrays of thousands of superconducting kinetic inductance detectors (KIDs) sensitive to radiation between 25 and 265 $\mu$m. Here, we present laboratory characterization of prototype, 25 -- 80 $\mu$m wavelength, low-volume, aluminum KIDs designed for the low-background environment expected with PRIMA. A compact parallel plate capacitor is used to minimize the detector footprint and suppress TLS noise. A novel resonant absorber is designed to enhance response in the band of interest. We present noise and optical efficiency measurements of these detectors taken with a low-background cryostat and a cryogenic blackbody. A microlens-hybridized KID array is found to be photon noise limited down to about 50 aW with a limiting detector NEP of about $6.5 \times 10^{-19}~\textrm{W/Hz}^{1/2}$. A fit to an NEP model shows that our optical system is well characterized and understood down to 50 aW. We discuss future plans for low-volume aluminum KID array development as well as the testbeds used for these measurements., Comment: 8 pages, 4 figures, 20th International Workshop on Low Temperature Detectors, submitted to the Journal of Low Temperature Physics