Sound source localization (SSL) under unknown or variable sound sources conditions remains a challenging task. Existing methods suffer from limitations such as grid resolution constraints, fixed output dimensionality and insufficient exploitation of mutual assitance between temporal and spatial information. In this paper, we propose an Envelope Separation Aided Multi-Task Learning model for blind source counting and localization, which adaptively generates attractors to estimate source numbers and jointly optimizes envelope separation and direction estimation through a multi-task learning model using permutation invariant training (PIT). Experimental results demonstrated that the proposed model achieved better performance, by leveraging temporal domain envelope separation to aid spatial localization, outperforming baseline approaches.
The evolution of Fe(II)-oxidizing microorganisms has been closely linked to the evolution of Earth's iron biogeochemical cycle and redox history. However, its impact on the coupled biogeochemical cycling of iron and phosphorus, particularly the distribution of iron-bound phosphate (PFe) in water columns, remains largely unexplored. This study elucidates the distinct Fe(II) oxidation mechanisms of the anoxygenic Rhodobacter ferrooxidans SW2 and the oxygenic Synechococcus sp. PCC 7002, along with the properties, transformation processes, and phosphate interactions of their biogenic iron (oxyhydr)oxides. SW2-mediated Fe(II) oxidation via iron oxidase drove sequential transformation from ferrihydrite to green rust and then to goethite. The resulting cell-mineral aggregates had a large hydrodynamic diameter (Dh, up to 26 μm), a high Fe/C ratio (∼2.5), and a rapid sedimentation rate (up to 57.7 m/day), efficiently transporting PFe to deep-sea sediments. In contrast, PCC 7002 indirectly oxidized Fe(II) via oxygen production, forming poorly crystalline iron (oxyhydr)oxides stabilized by extracellular polymeric substances. The resultant small aggregates (Dh = ∼6.9 μm), with a slower sedimentation rate (∼3.9 m/day), exhibited high phosphorus retention and were susceptible to dissimilatory iron reduction, facilitating PFe recycling in surface waters. These findings suggest that biogenic iron (oxyhydr)oxides from anoxygenic iron oxidizers act as carriers, transporting phosphorus to deep sediments, whereas those from oxygenic cyanobacteria function as phosphorus traps in surface waters. This study provides new insights into how the evolution of Fe(II)-oxidizing microorganisms reshapes PFe cycling and distribution in water columns, emphasizing the need to integrate microbiological and geochemical perspectives in understanding Earth's biogeochemical cycles.
Earth’s core has long been speculated to be the largest reservoir of hydrogen (H) on the planet. However, current estimates of its H content involve substantial uncertainties, due to the challenge of quantifying H under extreme conditions. Here, we perform superliquidus metal-silicate partitioning experiments on H using laser-heated diamond anvil cells, and combine it with atom probe tomography. The direct observation of H at silicon- and oxygen-rich nanostructures in the iron alloy indicates coupled sequestration of silicon, oxygen and hydrogen into Earth’s core during its formation. With the observed molar Si/H ratio close to unity, Earth’s core is estimated to contain 0.07-0.36 wt.% H, equivalent to 9-45 oceans of water. Such an amount would require the Earth to obtain the majority of its water from the main stages of terrestrial accretion, instead of through comets during late addition.
For biointegrated flexible systems that acquire and process electrophysiological signals, amplifying weak biosignals from their original low amplitudes (ranging from microvolt to millivolt) to volt-level is essential for subsequent processing. Achieving this level of amplification requires a high voltage gain (>105 or 100 decibels for microvolt signals). However, realizing such gain in flexible circuits remains highly challenging because of constrained integration scale and limits in feasible circuit topologies. Here, we report flexible amplifiers that achieve ultrahigh gain by leveraging intrinsic gain singularities induced by negative differential resistance (NDR) effect in carbon nanotube–based transistors. The NDR behavior is investigated under various factors, including contacts, gate structures, and channel lengths. Guided by insights into the correlations between NDR characteristics and device-level parameters, a device-circuit codesign approach is implemented to build a flexible amplifier achieving a record-high gain of 104 decibels among all reported flexible amplifiers, with successful demonstration of electroencephalogram signals amplification. A carbon nanotube–based flexible amplifier achieves >100-dB voltage gain, demonstrating EEG signal amplification.
The development of the sixth generation of wireless communications technology requires terminals that can operate at frequencies above 100 GHz. For human-centric applications, these terminals should also be flexible and have low power. However, current flexible radio-frequency transistors typically have lower maximum frequencies, in part due to the poor thermal conductivity of flexible substrates. Here we report radio-frequency transistors that are based on aligned carbon nanotube arrays on flexible substrates, having current-gain cut-off frequencies (fT) and power-gain cut-off frequencies (fmax) above 100 GHz. This is achieved by using electrothermal co-design to improve the heat dissipation and radio-frequency performance of the devices. The transistors exhibit an on-state current of 0.947 mA µm−1, a transconductance of 0.728 mS µm−1, a peak extrinsic fT of 152 GHz, a peak extrinsic fmax of 102 GHz and a power consumption under 200 mW mm−1. We also show that the devices can be used to create flexible radio-frequency amplifiers with an output power of 64 mW mm−1 and an 11-dB power gain in the K band.
Higher-Order Ambisonics (HOA) encoding from sparse, irregular microphone arrays remains a critical challenge for consumer spatial audio capture in immersive communication and XR. We propose Flow-HOA, a generative framework that jointly optimizes a multi-dimensional objective encompassing time-domain, spectral, and spatial fidelity while producing a deployable, time-invariant bank of Finite Impulse Response (FIR) encoding filters. Using conditional flow matching, the model learns to map a simple prior distribution to the target distribution of FIR filtercoefficients. Training is guided by a composite loss that balances time-domain waveform fidelity, multi-resolution spectral consistency, sub-band energy preservation, and spatial directivity constraints. Objective evaluations on synthetically simulated data demonstrate improved performance over strong model-based baselines in both signal fidelity and spatial accuracy metrics. Subjective listening tests on real microphone array recordings further confirmthat Flow-HOA yields higher overall sound quality with reduced artifacts, demonstrating generalization from synthetic training data to real-world capture conditions.
Chinese universities have actively engaged in knowledge production about other countries and played crucial roles in training professionals for the country’s foreign policy community since the founding of the People’s Republic (PRC). For decades, the transformations of the PRC’s international and area studies programs have not only reflected the shifting demands of top decision-makers but also influenced the interactions among scholars, academic institutions, government agencies, business elites, and the general public. This article analyzes the ups and downs of China’s international and area studies programs in the second half of the 20th century and how they were closely intertwined with the country’s shifting priorities in foreign policy, overseas Chinese affairs, and higher education. Primarily focusing on the case of Southeast Asian Studies (SEAS), this paper explores how the PRC’s area studies programs interacted with the country’s foreign policy during the 1955–1965 and post-1978 periods. Despite the changing geopolitical dynamics after 2000, the academic infrastructure and particular institutional culture formed in these two phases remain essential to our understanding of China’s area studies initiatives today.
This paper revisits the commercial peace hypothesis using global event data from 1989 to 2019. We show that greater imports consistently worsen bilateral relations, with the effect strengthening after the 2008 financial crisis. The negative impact is most pronounced in democracies, high-income countries and those with elevated unemployment, where globalisation's distributive conflicts more easily shape foreign policy. Instrumental variable estimates confirm causality, and analyses using militarised interstate disputes reveal that trade-driven tensions extend beyond diplomacy to overt conflict. These findings challenge the optimistic view that economic interdependence fosters peace.
Individualized head-related transfer functions (HRTFs) require accurate pinna geometry, yet commodity multi-view captures leave the ear region self-occluded and weakly textured. We present a practical pipeline that couples ear-centric acquisition with 3D Gaussian splatting (3DGS) and the boundary element method (BEM) for complete HRTF computation. The protocol augments horizontal views with per-ear elevated captures under directional lighting; 3DGS training with depth-distortion regularization yields watertight meshes via truncated signed distance function (TSDF) fusion. Standardized head coordinates and ear-canal annotations interface the mesh with BEM. Experimental evaluations demonstrate that our method achieves lower ear-region geometric error and lowerfull-band spectral distortion compared to existing image-based personalized reconstruction baselines including AudioEar, NeuS, and Metashape MVS.