arXiv subjects

Yiru Zhou

Publications and source records attributed to Yiru Zhou.

7 recordsLinked to original sources

arXiv2026

Metropolitan entanglement distribution between an atom and a near-visible photon

Entanglement distribution is the overarching purpose of quantum networks. While communication over long distances can use deployed fiber infrastructure, it requires photons in the telecom band. However, advanced quantum network nodes do not operate at such wavelengths. Here we overcome this limitation with two tailor-made low-noise quantum-frequency converters to distribute entanglement between a single atom and a resonant photon over 14km line-of-sight via 24km of a deployed commercial fiber. The photon at wavelength 780nm is first entangled with the atom, then converted to the telecom S-band, and finally back-converted after propagation through the fiber. This link enables a photon transfer efficiency of 1.7% while affecting the atom-photon entanglement fidelity by less than 1%. This brings integration of atomic quantum nodes with existing long-distance fiber networks into reach, enabling novel applications in quantum information processing.

quant-ph↗

arXiv2025

An Analysis of Large Language Models for Simulating User Responses in Surveys

Using Large Language Models (LLMs) to simulate user opinions has received growing attention. Yet LLMs, especially trained with reinforcement learning from human feedback (RLHF), are known to exhibit biases toward dominant viewpoints, raising concerns about their ability to represent users from diverse demographic and cultural backgrounds. In this work, we examine the extent to which LLMs can simulate human responses to cross-domain survey questions through direct prompting and chain-of-thought prompting. We further propose a claim diversification method CLAIMSIM, which elicits viewpoints from LLM parametric knowledge as contextual input. Experiments on the survey question answering task indicate that, while CLAIMSIM produces more diverse responses, both approaches struggle to accurately simulate users. Further analysis reveals two key limitations: (1) LLMs tend to maintain fixed viewpoints across varying demographic features, and generate single-perspective claims; and (2) when presented with conflicting claims, LLMs struggle to reason over nuanced differences among demographic features, limiting their ability to adapt responses to specific user profiles.

cs.CL↗

arXiv2024

SHMC-Net: A Mask-guided Feature Fusion Network for Sperm Head Morphology Classification

Male infertility accounts for about one-third of global infertility cases. Manual assessment of sperm abnormalities through head morphology analysis encounters issues of observer variability and diagnostic discrepancies among experts. Its alternative, Computer-Assisted Semen Analysis (CASA), suffers from low-quality sperm images, small datasets, and noisy class labels. We propose a new approach for sperm head morphology classification, called SHMC-Net, which uses segmentation masks of sperm heads to guide the morphology classification of sperm images. SHMC-Net generates reliable segmentation masks using image priors, refines object boundaries with an efficient graph-based method, and trains an image network with sperm head crops and a mask network with the corresponding masks. In the intermediate stages of the networks, image and mask features are fused with a fusion scheme to better learn morphological features. To handle noisy class labels and regularize training on small datasets, SHMC-Net applies Soft Mixup to combine mixup augmentation and a loss function. We achieve state-of-the-art results on SCIAN and HuSHeM datasets, outperforming methods that use additional pre-training or costly ensembling techniques.

cs.CV↗

arXiv2023

Long-lived quantum memory enabling atom-photon entanglement over 101 km telecom fiber

Long-distance entanglement distribution is the key task for quantum networks, enabling applications such as secure communication and distributed quantum computing. Here we report on novel developments extending the reach for sharing entanglement between a single $^{87}$Rb atom and a single photon over long optical fibers. To maintain a high fidelity during the long flight times through such fibers, the coherence time of the single atom is prolonged to 7 ms by applying a long-lived qubit encoding. In addition, the attenuation in the fibers is minimized by converting the photon's wavelength to the telecom S-Band via polarization-preserving quantum frequency conversion. This enables to observe entanglement between the atomic quantum memory and the emitted photon after passing 101 km of optical fiber with a fidelity better than 70.8$\pm$2.4%. The fidelity, however, is no longer reduced due to loss of coherence of the atom or photon but in the current setup rather due to detector dark counts, showing the suitability of our platform to realize city-to-city scale quantum network links.

quant-ph↗

arXiv2022

Improving Human Sperm Head Morphology Classification with Unsupervised Anatomical Feature Distillation

With rising male infertility, sperm head morphology classification becomes critical for accurate and timely clinical diagnosis. Recent deep learning (DL) morphology analysis methods achieve promising benchmark results, but leave performance and robustness on the table by relying on limited and possibly noisy class labels. To address this, we introduce a new DL training framework that leverages anatomical and image priors from human sperm microscopy crops to extract useful features without additional labeling cost. Our core idea is to distill sperm head information with reliably-generated pseudo-masks and unsupervised spatial prediction tasks. The predicted foreground masks from this distillation step are then leveraged to regularize and reduce image and label noise in the tuning stage. We evaluate our new approach on two public sperm datasets and achieve state-of-the-art performances (e.g. 65.9% SCIAN accuracy and 96.5% HuSHeM accuracy).

cs.CV↗

arXiv2021

Entangling single atoms over 33 km telecom fibre

Heralded entanglement between distant quantum memories is the key resource for quantum networks. Based on quantum repeater protocols, these networks will facilitate efficient large-scale quantum communication and distributed quantum computing. However, despite vast efforts, long-distance fibre based network links have not been realized yet. Here we present results demonstrating heralded entanglement between two independent, remote single-atom quantum memories generated over fibre links with a total length up to 33 km. To overcome the attenuation losses in the long optical fibres of photons initially emitted by the Rubidium quantum memories, we employ polarization-preserving quantum frequency conversion to the low loss telecom band. The presented work represents a milestone towards the realization of efficient quantum network links.

quant-ph↗

arXiv2019

Einstein-Podolsky-Rosen Energy-Time Entanglement of Narrowband Biphotons

We report the direct characterization of energy-time entanglement of narrowband biphotons produced from spontaneous four-wave mixing in cold atoms. The Stokes and anti-Stokes two-photon temporal correlation is measured by single-photon counters with nano second temporal resolution, and their joint spectrum is determined by using a narrow linewidth optical cavity. The energy-time entanglement is verified by the joint frequency-time uncertainty product of 0.063 +/- 0.0044, which does not only violate the separability criterion but also satisfies the continuous variable Einstein-Podolsky-Rosen steering inequality.

quant-ph↗