arXiv subjects

Jiuyi Zhang

Publications and source records attributed to Jiuyi Zhang.

5 recordsLinked to original sources

arXiv2026

Efficient Mode Conversion at 1064 nm via Bilayer Inverse Taper on Thin-Film Lithium Niobate

Efficient fiber-to-chip coupling remains challenging for thin-film lithium niobate (TFLN) devices operating at 1064 nm. Here, we fabricate and characterize an on-chip optical mode converter based on a bilayer inverse taper. The proposed structure demonstrated a coupling loss of 1.9 dB per facet for the TE mode, with a 1 dB bandwidth spanning 1055--1085 nm. The polarization-dependent alignment tolerance was measured along the lateral and vertical directions. Theoretical and simulation analyses indicate that, when using a lensed fiber with a mode field diameter of 2.5 $μ$m, the coupling loss can be reduced to as low as 0.48 dB per facet. These results demonstrate the feasibility of providing a high-efficiency solution for on-chip optical coupling at 1064 nm.

physics.optics↗

arXiv2026

Decoherence-induced Multiphoton Interference

Decoherence is usually deemed detrimental to quantum information processing. Its control and minimization require significant costs and operating overheads, constituting a major hurdle to commercialize quantum technology. Yet, quantum mechanics provides for counterintuitive, sometimes surprisingly useful, phenomena and effects associated with decoherence, leading to unusual practical utilities. Here we demonstrate such an example of fundamental interest and practical potential, where genuine quantum interference is created among multiple photons through their dissipative coupling to a shared reservoir. On a thin-film lithium niobate chip, we incoherently link two spontaneous parametric down-converters through a common, highly-lossy channel to create coherent multiphoton states. Our results show that faithful correlations can be established among two, three, and four photons, and tuned by shifting the relative phase between the driving pumps for the converters. This experiment highlights an under-explored territory in quantum science and technology, where loss and decoherence serve as resources, rather than adversaries, for quantum information processing.

quant-ph↗

arXiv2026

Systematic Analysis of Ferroelectric Domain Dynamics in Periodically Poled Lithium Niobate Waveguides Using Two-Photon Microscopy and Digital Imaging Processing

We present a characterization and analysis methodology suitable for volume production for characterizing and optimizing x-cut thin-film periodically poled lithium niobate (PPLN) devices using two-photon (2P) microscopy with quantitative image processing. This method enables direct extraction of key structural parameters, such as duty cycle, phase-matching behavior, and domain uniformity, across large device sets in a non-destructive manner. By correlating 2P microscopy-derived structural metrics with systematic variations in poling conditions, we establish a scalable, image-driven approach for evaluating and improving PPLN fabrication. The resulting workflow supports wafer-level process development and accelerates the manufacturing and packaging of lithium niobate photonic integrated circuits (PICs).

physics.optics↗

arXiv2025

Filtering with Time-frequency Analysis: An Adaptive and Lightweight Model for Sequential Recommender Systems Based on Discrete Wavelet Transform

Sequential Recommender Systems (SRS) aim to model sequential behaviors of users to capture their interests which usually evolve over time. Transformer-based SRS have achieved distinguished successes recently. However, studies reveal self-attention mechanism in Transformer-based models is essentially a low-pass filter and ignores high frequency information potentially including meaningful user interest patterns. This motivates us to seek better filtering technologies for SRS, and finally we find Discrete Wavelet Transform (DWT), a famous time-frequency analysis technique from digital signal processing field, can effectively process both low-frequency and high-frequency information. We design an adaptive time-frequency filter with DWT technique, which decomposes user interests into multiple signals with different frequency and time, and can automatically learn weights of these signals. Furthermore, we develop DWTRec, a model for sequential recommendation all based on the adaptive time-frequency filter. Thanks to fast DWT technique, DWTRec has a lower time complexity and space complexity theoretically, and is Proficient in modeling long sequences. Experiments show that our model outperforms state-of-the-art baseline models in datasets with different domains, sparsity levels and average sequence lengths. Especially, our model shows great performance increase in contrast with previous models when the sequence grows longer, which demonstrates another advantage of our model.

cs.IR↗

arXiv2024

Chip-integrated Spectroscopy Capable of Temperature Retrieval

We demonstrate a chip-integrated emission spectroscope capable of retrieving the temperature of the light sources. It consists of a single photon detector with low dark counts and a sweeping on-chip filter with 2 pm spectral resolution in the visible and near-infrared regimes. With wildfire sensing applications in mind, we test our system with a hollow cathode lamp to simulate the K-line emission, and show how the models of Doppler and collision broadening in the plasma can be used for temperature retrieval. With favorable device parameters, high spectral resolution, and a novel temperature retrieval capability, our technique may find broad applications in environmental monitoring, astrophysics, plasma physics, and so on.

physics.optics↗