Abstract Immersive applications call for synthesizing spatiotemporal 4D content from casual videos without costly 3D supervision. Existing video‐to‐4D methods typically rely on manually annotated camera poses, which are labor‐intensive and brittle for in‐the‐wild footage. Recent warp‐then‐inpaint approaches mitigate the need for pose labels by warping input frames along a novel camera trajectory and using an inpainting model to fill missing regions, thereby depicting the 4D scene from diverse viewpoints. However, this trajectory‐to‐trajectory formulation often entangles camera motion with scene dynamics and complicates both modeling and inference. We introduce S ee 4D , a pose‐free, trajectory‐to‐camera framework that replaces explicit trajectory prediction with rendering to a bank of fixed virtual cameras, thereby separating camera control from scene modeling. A view‐conditional video inpainting model is trained to learn a robust geometry prior by denoising realistically synthesized warped images and to inpaint occluded or missing regions across virtual viewpoints, eliminating the need for explicit 3D annotations. Building on this inpainting core, we design a spatiotemporal autoregressive inference pipeline that traverses virtual‐camera splines and extends videos with overlapping windows, enabling coherent generation at bounded per‐step complexity. We validate See4D on cross‐view video generation and sparse reconstruction benchmarks. Across quantitative metrics and qualitative assessments, our method achieves superior generalization and improved performance relative to pose‐ or trajectory‐conditioned baselines, advancing practical 4D world modeling from casual videos.

In a laser, the control of its spectral emission depends on the physical dimensions of the optical resonator, restricting it to a set of discrete cavity modes at specific frequencies1-4. Without modifying the optical cavity, this results in substantial gaps in the obtainable laser emission spectrum, as well as a fixed repetition rate, limiting the device's usability in various experiments and applications where a considerable degree of tunability is required in the spectral or temporal domain. Here we overcome this fundamental limit by demonstrating a monolithic semiconductor laser5-7 with a continuously tunable repetition rate from 4 GHz up to 16 GHz, by using a microwave driving signal that induces a spatiotemporal gain modulation along the entire laser cavity8,9, generating intracavity mode-locked pulses10-13 with a continuously tunable group velocity14. At the output, frequency combs15,16 with continuously tunable mode spacings are generated in the frequency domain, and coherent pulse trains with continuously tunable repetition rates are generated in the time domain17. Our results pave the way for fully tunable chip-scale lasers and frequency combs, which will be advantageous for use in a diverse variety of fields, from fundamental studies to applications such as high-resolution and dual-comb spectroscopy18,19.

Abstract Reconstructing detailed geometry and realistic appearance from a single RGB image is essential yet fundamentally challenging due to inherent ambiguities such as occlusion, lighting variations, and texture‐geometry entanglement. While recent diffusion‐based generative models have significantly improved novel view synthesis, existing approaches suffer from two critical limitations: lack of cross‐view geometric consistency and insufficient cross‐domain semantic alignment. To address these issues, we introduce U ni C ross 3D , a unified cross‐view and cross‐domain diffusion framework designed explicitly for consistent and physically coherent 3D generation. U ni C ross 3D features two novel contributions: (1) a cross‐view latent regularization that enforces cross‐view geometric consistency across synthesized viewpoints by penalizing latent variance, and (2) a cross‐domain mutual information objective grounded in the physics of image formation, explicitly aligning synthesized color and normal maps. Extensive experiments demonstrate that U ni C ross 3D achieves significantly improved view consistency and semantic alignment over state‐of‐the‐art methods and yields higher‐fidelity reconstructions, particularly under challenging textures and ambiguous viewpoints.

This paper presents the hardware-software implementation of a computerized system for predictive identification of parameters of a multilayer medium of the “oil–emulsion–water” type. The proposed approach is based on a combination of the active compensation method, digital signal synthesis, and synchronous detection. An adaptive transimpedance measurement path and methods for mathematical compensation of hardware errors are implemented. It is shown that the use of coherent signal generation and a ratiometric approach makes it possible to minimize the influence of parasitic parameters and ensure high measurement accuracy over a wide frequency range.

Generation of quantum coherence in a magnomechanical system

Integration of multimodal, multi-omics data is critical for advancing precision medicine, yet its application is frequently limited by incomplete datasets where one or more modalities are missing. To address this challenge, we developed a generative framework capable of synthesizing any missing modality from an arbitrary subset of available modalities. We introduce Coherent Denoising, a novel ensemble-based generative diffusion method that aggregates predictions from multiple specialized, single-condition models and enforces consensus during the sampling process. We compare this approach against a multi-condition, generative model that uses a flexible masking strategy to handle arbitrary subsets of inputs. The results show that our architectures successfully generate high-fidelity data that preserve the complex biological signals required for downstream tasks. We demonstrate that the generated synthetic data can be used to maintain the performance of predictive models on incomplete patient profiles and can leverage counterfactual analysis to guide the prioritization of diagnostic tests. We validated the framework's efficacy on a large-scale multimodal, multi-omics cohort from The Cancer Genome Atlas (TCGA) of over 10,000 samples spanning across 20 tumor types, using data modalities such as copy-number alterations (CNA), transcriptomics (RNA-Seq), proteomics (RPPA), and histopathology (WSI). This work establishes a robust and flexible generative framework to address sparsity in multimodal datasets, providing a key step toward improving precision oncology.

Abstract We introduce the Palette‐Adapter , a novel method for conditioning text‐to‐image diffusion models on a user‐specified color palette. While palettes are a compact and intuitive tool widely used in creative workflows, they introduce significant ambiguity and instability when used for conditioning image generation. Our approach addresses this challenge by interpreting palettes as sparse histograms and introducing two scalar control parameters: histogram entropy and palette‐to‐histogram distance , which allow flexible control over the degree of palette adherence and color variation. We further introduce a negative histogram mechanism that allows users to suppress specific undesired hues, improving adherence to the intended palette under the standard classifier‐free guidance mechanism. To ensure broad generalization across the color space, we train on a carefully curated dataset with balanced coverage of rare and common colors. Our method enables stable, semantically coherent generation across a wide range of palettes and prompts. We evaluate our method qualitatively, quantitatively, and through a human evaluation, and show that it consistently outperforms existing approaches in achieving both strong palette adherence and high image quality.

We present a polarisation-maintaining all-normal dispersion photonic crystal fibre designed for 1030nm femtosecond pumping, enabling ultra-stable and coherent supercontinuum (SC) generation spanning 650-1300nm. The fibre's polarisation-maintaining properties are achieved through two larger central holes in the structure, which is an alternative approach to using conventional stress rods. The fibre is specifically engineered to achieve minimum dispersion near 1030nm, making it ideal for ultrafast comb-based metrology, and widely tunable optical parametric amplifier (OPA) systems. We further investigate the influence of input pulse contrast on supercontinuum generation through both numerical simulations and experiments. Relative intensity noise (RIN) and phase noise (PN) are characterized using three complementary techniques: dispersive Fourier transform (DFT), the Bellini-Hänsch interferometric method, and the dual-reference oscillator cross-correlation technique. The results demonstrate excellent stability, with pulse-to-pulse RIN below 0.5%, an optical phase deviation under 15mrad, and phase noise levels down to - 150dBc/Hz at 10kHz from the carrier, confirming the fibre's suitability for demanding ultrafast applications.

The dynamics of trade-off relations between quantum information resources are investigated in a system of two non-interacting qubits, each individually and nonlinearly coupled to coherent cavity fields through intensity-dependent coupling. The qubits are initially prepared in a maximally entangled Bell state, while the cavity fields are assumed to be in coherent states. An exact analytical treatment of the reduced two-qubit density matrix is presented under qubit–cavity detuning and intrinsic decoherence governed by the Milburn model. The time evolutions of first-order coherence, concurrence, intrinsic concurrence and purity are analyzed to elucidate the redistribution of quantum resources. The results confirm that intrinsic concurrence and first-order coherence obey a complementary trade-off relation governed by the purity of the two-qubit state, with standard concurrence acting as a lower bound for intrinsic concurrence throughout the dynamics. The roles of the initial coherent field intensity, qubit–cavity detuning and intrinsic decoherence are examined, showing that nonlinear atom–cavity interactions induce regular oscillatory behavior accompanied by entanglement sudden death and revival phenomena. Increasing the coherent field intensity enhances coherence generation and accelerates entanglement degradation of the generated two-qubit states.

We demonstrate coherent orbital angular momentum (OAM) generation from chaotic phase screens with discrete integer-valued azimuthal bias. Through Fourier analysis, we prove that coherent OAM amplitude is nonzero only when the ensemble-averaged complex phase factor has corresponding integer Fourier coefficients. For discretized Gaussian bias, the coherent power follows a universal exponential profile in scaled coordinates, while continuous bias yields sinc-filtered spectra with algebraic decay. Monte Carlo simulations validate these selection rules, showing forbidden level suppression exceeding four orders of magnitude and correlation with theory above 99.9 percent. The discrete statistical structure enables deterministic control of coherent OAM content, with immediate applications in optical communications and sensing via spatial light modulators or metasurfaces. Beyond static scalar beams, we show that coherent selection extends to spin-orbit coupled vector beams and enables time-multiplexed coherent filters via dynamic discrete bias, both of which elevate practical utility without additional hardware complexity.

Abstract The most general searches for gravitational wave transients (GWTs) rely on data analysis methods that do not assume prior knowledge of the signal’s waveform, direction, and arrival time on Earth. These searches provide data-driven signal reconstructions that are crucial both for testing available GW emission models and for discovering yet-to-be-unveiled sources. Here, we discuss the progresses in detection performances of the coherent WaveBurst second generation pipeline (cWB-2G), which is highly adaptable to minimally–modeled as well as model–informed searches for GWTs. Several search configurations for GWTs are examined using approximately 14.8 d of observation time from the third observing run by LIGO-Virgo-KAGRA (LVK). Recent enhancements include a ranking statistic fully based on multivariate classification with eXtreme gradient boosting, a thorough validation of the accuracy of the statistical significance of GWT candidates, and the measurement of the correlations of false alarms and simulated detections between different concurrent searches. For the first time, we provide a comprehensive comparison of cWB-2G performances on data from networks made of two and three detectors, and we demonstrate the advantage of combining concurrent searches for GWTs of generic morphology in a global observatory. This work offers essential insights for assessing our data analysis strategies in the ongoing and future LVK searches for generic GWTs.

Electroencephalography (EEG) provides a non-invasive window into neural dynamics at high temporal resolution and plays a pivotal role in clinical neuroscience research. Despite this potential, prevailing computational approaches to EEG analysis remain largely confined to task-specific classification objectives or coarse-grained pattern recognition, offering limited support for clinically meaningful interpretation. To address these limitations, we introduce NeuroNarrator, the first generalist EEG-to-text foundation model designed to translate electrophysiological segments into precise clinical narratives. A cornerstone of this framework is the curation of NeuroCorpus-160K, the first harmonized largescale resource pairing over 160,000 EEG segments with structured, clinically grounded natural-language descriptions. Our architecture first aligns temporal EEG waveforms with spatial topographic maps via a rigorous contrastive objective, establishing spectro-spatially grounded representations. Building on this grounding, we condition a Large Language Model through a state-space-inspired formulation that integrates historical temporal and spectral context to support coherent clinical narrative generation. This approach establishes a principled bridge between continuous signal dynamics and discrete clinical language, enabling interpretable narrative generation that facilitates expert interpretation and supports clinical reporting workflows. Extensive evaluations across diverse benchmarks and zero-shot transfer tasks highlight NeuroNarrator's capacity to integrate temporal, spectral, and spatial dynamics, positioning it as a foundational framework for time-frequency-aware, open-ended clinical interpretation of electrophysiological data.

ABSTRACT We study the steady‐state quantum Fisher information (QFI) and several quantum resources entanglement, quantum coherence, and quantum discord for a pair of coupled qubits interacting with either bosonic or fermionic reservoirs. Using the Bloch–Redfield master equation beyond the secular approximation, we derive analytical expressions for the steady‐state density matrix and analyze how equilibrium and nonequilibrium conditions shape the resulting quantum correlations. In thermal equilibrium, the quantum resources display distinct behaviors depending on the bath statistics, with fermionic reservoirs enabling stronger stationary correlations due to particle‐exchange processes. Under nonequilibrium driving, generated either by a temperature imbalance or by a chemical potential difference, the system develops steady‐state coherence and enhanced discord, while entanglement and QFI exhibit non‐monotonic dependence governed by the interplay between coherence generation and population mixing. Our results identify the regimes in which steady‐state metrological performance can be optimized, and they provide general guidelines for enhancing quantum correlations in dissipative two‐qubit platforms.

Reconfigurable coherent frequency-stepped signal generation based on recirculating optoelectronic spectrum expansion

Copper selenide is an exceptional quasi-layered monolithic material that exhibits both semiconducting and metallic properties in adjacent visible and near-infrared (NIR) spectral ranges. Here we introduce a thiol-free colloidal synthesis for generating quasi-2D klockmannite copper selenide nanocrystals via hot injection method, achieving shape control by tuning the injection temperature and precursor concentrations without any additional ligands. This approach produces large klockmannite nanosheets with lateral sizes from 200nm to several micrometers, as well as uniform triangular nanoplatelets with sizes of 12-25nm that are monocrystalline and display strong NIR plasmonic absorption. The spectral features of the anisotropic klockmannite phase in the NIR have been analyzed using complex-scaled discrete dipole approximation (CSDDA) calculations, which reveal pronounced optical anisotropy and the emergence of hyperbolic regime. The combined effect of propagating and evanescent fields is regarded as the underlying reason of such modes in the hyperbolic domain. Finally, the ultrafast photophysical behavior of the material in klockmannite phase is examined, including hot-hole cooling, trapping, and coherent phonons generation. Our findings emphasize the important role of the intrinsic crystal anisotropy in governing the physical properties of nanoscale klockmannite.

This work introduces a dispersion engineering technology for supercontinuum (SC) generation based on liquid-infiltrated suspended core fibers (SCFs), enabling efficient spectral control without modifying the fiber geometry. An SCF structure with air holes is filled with water, demonstrating a flat anomalous dispersion in a wide wavelength range. For coherent SC generation, an ethanol-filled fiber with the same core diameter of 1.7µm is considered to shift the dispersion curves toward the normal dispersion region. The resulting dispersion profiles show good agreement between numerical simulations and experimental data. The generalized nonlinear Schrödinger equation (GNLSE) is subsequently employed to investigate the influence of pump pulse parameters on SC evolution. The simulations investigate how SC changes with pump peak power and pulse width in a 10cm fiber when pumped at a wavelength of 1030nm. A broadening of the SC spectrum is observed as the pump peak power at a wavelength of 1030nm rises from 5 to 30kW. The same trend persists for shorter pulse widths, ranging from 200 to 50fs. Notably, the proposed water-filled SCF generates a broad spectrum spanning from 674 to 1918nm in the anomalous regimes. Meanwhile, the generation of a flat SC spectrum from 757 to 1400nm is obtained in the all-normal dispersion ethanol-filled fiber. These findings demonstrate the effectiveness of liquid infiltration as an experimentally accessible tool for dispersion control in SCFs, and provide new insights for infrared SC sources.

Integrated green light sources are essential for telecommunications and quantum applications, while the performance of current on-chip green light generation is still limited in power and tunability. In this work, we demonstrate green light generation in silicon nitride microresonators using photo-induced second-order nonlinearities, achieving up to 3.5 mW green power via second-harmonic generation and densely tunable over a 29 nm range. In addition, we report milliwatt-level all-optical poling (AOP) threshold, allowing for amplifier-free continuous-wave AOP. Furthermore, we demonstrate non-cascaded sum-frequency generation, leveraging the combination of AOP and simultaneous coherent frequency combs generation at 1 μm. Such comb-assisted AOP enables switching of the green light generation over an 11 nm range while maintaining the pump within a single resonance. The combination of such highly efficient photo-induced nonlinearity and multi-wavelength AOP enables the realization of low-threshold, high-power, widely-tunable on-chip green sources.

Frequency combs represent exceptionally precise measurement tools due to the coherence of their spectral lines. While optical frequency comb sources constitute a well-established technology, superconducting circuits provide a relatively unexplored on-chip platform for low-dissipation comb emitters able to span from gigahertz to terahertz frequencies. We demonstrate coherent microwave frequency comb generation by leveraging the ac Josephson effect in a superconducting quantum interference device. A time-dependent magnetic drive periodically generates voltage pulses, which in the frequency domain correspond to a comb with dozens of spectral modes here reported up to mode 46. The emitted power at the device level ranges from -170 dBm to -130 dBm per harmonic, corresponding to 40 dB dynamic range in the 4-8 GHz bandwidth. The micrometer-scale footprint and minimal dissipation inherent to superconducting systems foster the integration of our comb generator with advanced cryogenic electronics. Transferring optical techniques to the solid-state domain may enable new applications in quantum technologies.

Laser frequency conversion in wide bandgap nonlinear optical thin films enables compact coherent ultraviolet (UV) light sources for UV photonics integrated circuits. This work unveils the generation of coherent 350 nm wavelength UVA light in a-axis oriented wurtzite Zn0.67Mg0.33O and Zn0.51Mg0.49O thin films through frequency doubling of a femtosecond laser. The conversion efficiency of the 0.66 μm-thick Zn0.67Mg0.33O and 0.40 μm-thick Zn0.51Mg0.49O thin films reached 3.25 × 10−4 and 1.07 × 10−4, respectively. The single-phase wurtzite Zn0.51Mg0.49O thin film shows an optical bandgap of 4.13 eV, a light absorption coefficient of less than 1 cm−1 at 350 nm wavelength, and a second-order nonlinear susceptibility element χ33(2) of 10.83 pm/V at 700 nm wavelength. Increasing Mg incorporation into wurtzite ZnO causes lattice distortion and deteriorates the non-centrosymmetry of the wurtzite structure, resulting in the reduction in χ33(2). The introduction of a Zn0.79Mg0.21O buffer layer enhances the χ33(2) of the Zn0.51Mg0.49O thin film by 1.3 times through improving film growth quality and relieving the lattice distortion. These results suggest the great prospects of high-efficiency coherent UV light generation in widegap Zn1−xMgxO thin films integrated with reversible polarization.

Coherent generation and transfer of distant Bi- and tripartite entanglement in a hybrid atom-opto-magnomechanical system